Saudi Arabia’s Future Investment Initiative (FII) kicked off this week in Miami, and AI took center stage of the dialogue. I have been following the panel sessions for unique insights which inform where AI is going amidst the current war. I highly encourage readers to watch them on FII’s Youtube. The below readout is a quick summary of the points I have found interesting so far.

Subscribe now

The shadow of Iran’s war with the United States and Israel loomed over every session at FII Priority Miami this week. With the Strait of Hormuz effectively closed and regional energy disruptions now eclipsing the shocks of the COVID-19 era, the geopolitical stakes have reached a breaking point. Most significantly, the IRGC’s strikes against Amazon-owned data centers in the UAE and Bahrain mark the historical first military military assault on private-sector cloud infrastructure. Despite the regional chaos, Saudi Arabia is in Miami projecting aggressive continuity on its AI ambitions to an audience of investors, executives, policymakers, and world leaders. PIF Governor Yasir Al Rumayyan told the summit the Saudi fiscal position “remains strong, stable and resilient.” HUMAIN CEO Tareq Amin was blunter: “We will not be taking it easy in terms of our ambition.”

I wanted to share a few key takeaways after watching the sessions so far:

1. Saudi Arabia is now selling AI to the United States

The headline deal at FII Miami was HUMAIN’s partnership with Silicon Valley’s Turing to build what both companies describe as the world’s first enterprise AI agent marketplace, hosted on HUMAIN’s own platform, HUMAIN ONE. This is HUMAIN’s first US-based customer. Amin underscored the significance of this: “Who would have thought Saudi Arabia could export technology to the United States? Who would have thought this is possible?”

This is a major milestone for HUMAIN and Saudi’s AI ambitions. It signals one of the first major shifts from sovereign wealth funds in the Gulf writing checks to American technology companies to those American companies buying new digital products from a Gulf company. HUMAIN has successfully shown the capability for Gulf companies to invert that relationship. HUMAIN ONE will be a multi-agent orchestration platform that Amin described as an enterprise operating system with AI agents running core business functions like HR, finance, and legal, and with human employees acting as builders and supervisors, rather than operators. Turing will be HUMAIN’s first external customer deploying the new ONE system.

Why it matters: If HUMAIN can make the platform-exporter model work, it helps Saudi Arabia’s position in the global AI value chain shift from consumer to viable producer. It also helps balance the bilateral dependency that Saudi Arabia has deliberately cultivated with Washington and helps sell the long-term partnership value for linking American firms to Saudi compute and new technologies. Steven Cook wrote with Foreign Policy in February that Gulf states are investing in AI partly to guarantee American security commitments by making themselves essential to the US AI competition with China and they ensure their adversaries become Washington’s adversaries too. Selling platforms back to the US built on U.S. hardware (U.S. provided semiconductors and microchips) is the next iteration of that logic.

2. HUMAIN’s argument: organizational structure matters at this stage of AI development

The most substantive Gulf AI discussion at FII Miami was Amin’s appearance on a panel alongside IBM’s Gary Cohn and Riyadh Air CEO Tony Douglas. Amin opened by citing an MIT study finding that 95 percent of companies investing heavily in AI fail to see significant returns. He explained this was the result of corporate and organization structure, not hardware.

“If you approach models and AI applications in the format of just deploying it on top of legacy systems, legacy process, legacy organization, I will tell you you will not deliver on the result you want,”

HUMAIN, he argued, avoided this trap because it was built from scratch with no legacy ERP, no inherited workflows, and no organizational silos. Instead, their new product HUMAIN ONE was built as its own operating layer from day one. The company runs a revenue-share model for employees who build products that sell externally. This, he says, can turn any English-speaking thinker into a builder without exposure or deep knowledge of coding. He argued that what takes four months on a typical software development life cycle can now be done in may two to four days.

IMB’s Cohn offered the counterpoint from legacy enterprise. Most of the world’s GDP he argued is built on regulated industries like banking, healthcare, airlines, utilities, and regulators do not care if a company is mid-transformation toward a more AI-integrated model. They care that you are compliant. HE also pushed back on the 95 percent failure rate, arguing it reflects a reporting gap. Technologists and engineers inside large companies are using AI daily, but not telling leadership, because they are protecting their jobs.

Why it matters: Amin argues for Saudi Arabia’s greenfield advantage as an AI adoption advantage. From his view, since HUMAIN has no legacy systems, no inherited bureaucracy, no organizational debt, the invitation to global enterprise goes something like: “build on our platform because we already operate the way you wish you could.” In practice, HUMAIN is only twelve months old and is beginning to show what it is capable of delivering. But unleashing virtually unlimited capital to a single entity with top-down objective of translating Saudi AI ambitions into reality is a tall order. Time will tell if it can deliver at scale, especially amidst the current conflict and the post-conflict economic landscape. Investors at FII seemed bullish on its trajectory, and the regional chaos may also give them a distinct advantage given Saudi’s geography and economy make it more resilient to shocks than other Gulf countries. Investors may see Riyadh as a good medium term alternative to other gulf economies which are sustaining worse damage as a result of the war.

3. PIF is shifting from sole funder to platform operator

Al Rumayyan used FII Miami to announce that PIF’s new five-year strategy (2026–2030) will prioritize attracting third-party capital into PIF portfolio companies rather than funding everything with sovereign equity. “We want to get more and more people to work with us, to encourage third-party capital to work with us,” he told the summit. This is a structural departure from the fund’s previous model of leading massive projects with its own capital.

Why it matters: For AI specifically, this means HUMAIN and other PIF-backed technology companies will increasingly need to compete to attract commercial investment on their own merits, not just sovereign backing.

4. Latin America is emerging as a new corridor for Saudi AI and technology capital

Day 0 of FII Miami was entirely dedicated to a “New Latam Order Summit,” and positioned the region as a strategic investment destination for Gulf capital. Saudi Eksab and BTG Pactual signed a framework agreement to create a Latin America-focused alternative investment platform. Eksab also signed a separate agreement with the Inter-American Development Bank Group to build a joint pipeline of direct and indirect investment opportunities, initially focused on Central America and the Caribbean, with plans for a co-investment vehicle backed by IDB Invest. Eksab CEO Yazeed Alyahya said AI is the centerppiece of the strategy, telling Arab News that Crown Prince Mohammed bin Salman designated 2026 the year of artificial intelligence. FII CEO Richard Attias put it bluntly: “Latin America is no longer a region of potential. It is a region of decision.”

Why it matters: Saudi sovereign capital is diversifying its investments globally, and has set its sights on Latin America as a deployment destination. The sheer number of deals announced around Gulf-LATAM times shows that region is set to play an increasingly important role in the Gulf investment picture, and investors are building the institutional infrastructure to channel that capital in real time. One unique question I have is whether Gulf investment will favor U.S. or Chinese influence and investments in Latin America. After Trump’s Venezuela invasion, the future of Chinese investment has been somewhat debated, but the Gulf may be an acceptable backstop for Latin American countries which want a third option. This was the assertion of my previous paper (see below) and seems to reinforcement my initial assessment about the GCC becomes the “third" option for the global south.

5. Discussion on attacks on data centers as military targets was noticeably absent

The AI panels at FII Miami acknowledged the Iran crisis, but discussions on the Iranian strikes on AI infrastructure was noticeably absent from the conversation. The panel format allowed the Saudi event to project normalcy on the technology track while the geopolitical crisis played out elsewhere in the summit. Investors and hyperscalers, more generally, seem to be treating the strikes as an isolated episode, not a structural disqualifier for investments in the Gulf. No major firm has announced a withdrawal from the region. The HUMAIN-Turing deal went ahead on schedule (though this is not an infrastructure deal, it shows that US firms are still willing to do business). The working assumption among Gulf policymakers appears to be that physical risk can be mitigated — through hardened facilities, geographic diversification, and potentially some layered air defense systems.

Concluding Thoughts

Silicon Valley and U.S. investors remain notably bullish on Saudi Arabia despite regional instability. While high-level summits are rarely perfect barometers of sentiment, the sustained enthusiasm for KSA’s HUMAIN initiative and broader positive energy around the Kingdom’s deployment of AI suggests that investor confidence is not in retreat. Beyond structural incentives like subsidized power and land, Riyadh may emerge as a ‘safe harbor’ for AI capital. As neighboring Gulf states with competitive industries face the strain of Iranian-backed escalations, their focus has necessarily shifted to national defense, economic preservation, and protection of their populations. In contrast, Saudi Arabia’s ability to remain largely unscathed has allowed it to maintain its AI development momentum. This regional divergence could provide Riyadh with the strategic window it needs to close the gap created by the UAE’s early lead in the AI sector.

Thanks for reading. Make sure to subscribe for more.

Subscribe now

List of videos:

• FII Priority MIAMI 2026: How to Go AI-Native
  

• FII Priority MIAMI 2026: Where Is AI’s ROI?
  

• FII Priority MIAMI 2026: Is the Hyperscaler Partnership the Future of AI?
  

• FII Priority MIAMI 2026 DAY 2: Which Compute Hubs Will Win?
  

• FII PRIORITY Miami 2026 | Opening Show
  

• FII Priority MIAMI 2026: Building the Infrastructure for the On-Chain and AI Economy
  

Washington and Beijing share core interests in preventing a nuclear Iran and securing Gulf energy flows, yet diverge sharply on force, sanctions, and regime change, limiting coordination despite rising costs of inaction.

“The cost of non-coordination on the Gulf has risen high enough that limited alignment now looks more attractive than avoiding it. Neither Washington nor Beijing will reconcile their theories of order, but both now pay a visible price for the lack of coordination. Trump has demanded Chinese help in keeping the Strait of Hormuz open, even though the closure results from a war China considers illegal. Trump’s claim that China receives ninety percent of its oil through the Strait is dramatically overstated, but it frames non-cooperation as free-riding. The Bessent–He Lifeng trade talks in Paris suggest both sides know cancellation is costly. The delay is leverage, not policy. It signals to Beijing that Washington may view the summit more as a bargaining tool than as a space for dialogue.

Beijing’s rational response is patience and silence. China is unlikely to send ships, pressure Tehran, or broker terms of a deal. Beijing would rather absorb short-term disruption than deliver Washington a win that validates the use of force or justifies regime change. Beijing’s official response exemplifies this careful balancing act: “As a sincere friend and strategic partner of Middle Eastern countries, China will continue to strengthen communication with relevant parties, including parties to the conflict, and play a constructive role for de-escalation and restoration of peace.”

Subscribe now

This is the next in a serise of ongoing surveys of Chinese-language sources on how Operation Epic Fury is reshaping the debate over compute infrastructure, AI militarization, and the future of tech investment in the GCC. This update provides a couple of other voices I missed on March 1-2 (which had interesting views worth sharing, as well as more recent commentary from the past few days. I conclude with analysis of whether the current moment is a business opportunity for Chinese hyperscalers and how this impacts Gulf digital sovereignty.

Subscribe now

Since Iran’s retaliatory missile and drone strikes began hitting Gulf Arab states on March 1, Chinese scholars, analysts, and financial commentators have produced a body of commentary which helps us pinpoint how Beijing thinkers are evaluating the unfolding conflict. Beijing’s analysts this week are asking a an import set of questions about who controls the nerve center of the AI era, about what the war reveals about American technological dependency, and about whether the conflict opens a door for Chinese compute providers to displace their American competitors across the Global South.

Lets dig in.

Data Centers as Strategic Targets

I found the most analytically precise Chinese framing of the AWS strikes came from Sun Chenghao, a researcher at Tsinghua University’s Center for Strategic and Security Studies and author of the popular substack (make sure to subscribe). Writing in First Financial (republished on Guancha) on March 6, Sun argued that data centers are approaching the status of a new class of strategic target. He argues that in the industrial era, military strikes aimed at oil and gas facilities, power plants, ports, and communications hubs disrupt the “blood supply system” of the economy. In the age of AI and cloud computing, he said, compute and data infrastructure have become a nation’s “nerve center.”

Sun stressed that the vulnerability is asymmetric. An adversary does not need to destroy an entire facility. Disrupting power supply, cooling systems, or key network nodes alone can cause prolonged outages that cascade into financial, logistics, and other civilian systems. The Middle East now hosts more than 300 data centers, with over half concentrated in Israel, Saudi Arabia, and the UAE — a density that Iran has demonstrated it can reach.

I would be eager to hear ’s view on risks of Chinese hyperscalers and tech companies in the GCC countries.

Zhang Gaosheng, a scholar at the CIIS World Peace and Security Institute, extended this argument in a March 16 Global Times interview. He warned that Western commercial cloud-stored data — satellite imagery, communications intercepts, financial transaction records, social media traces — can easily be accessed by the US military in wartime and converted into targeting intelligence. In his policy conclusion, he argues that data centers, compute nodes, and communications cables now directly determine national security and wartime initiative. China, he argued, must accelerate the construction of autonomous and controllable cloud infrastructure, push for data localization, and build backup mechanisms to ensure survivability under extreme conditions.

The phrase Zhang used — "independent and controllable" (自主可控) — is worth pausing on. 自主 carries the sense of self-directed, acting on one's own authority without external dependence; 可控 means controllable. In this context Zhang is referencing national sovereignty and agency over technology supply chains. China controlling its own stack, rather than depending on American infrastructure that can be sanctioned, weaponized, or, as the war has now demonstrated, physically destroyed. It is the standard PRC policy vocabulary for indigenous technology substitution, the same formulation that runs through Made in China 2025 and Beijing's semiconductor self-sufficiency drive. Interestingly, Zhang’s writing carries a bit of deliberate ambiguity. Zhang may be directing his message inward, telling Chinese enterprises operating in the Gulf to migrate off American cloud services and onto domestic platforms for their own resilience. He may also be signaling outward, that Huawei Cloud, Alibaba Cloud, and other Chinese providers should move into the markets the conflict is destabilizing to provide sustainable alternatives to U.S. hyperscalers under attack. What Zhang does not say explicitly, another Chinese analyst states in plain commercial terms (see toward the end of the report): Iran’s AWS strikes validate an export play for Chinese compute solutions across the Middle East. In short, China’s commercial selling point to the Middle East is “our data centers won’t be bombed by Iran.”

AI in the Kill Chain: Chinese Coverage of Claude’s Battlefield Role

Chinese media covered the use of AI in Operation Epic Fury with considerable depth and, in some cases, a level of granularity that exceeded mainstream Western reporting.

The most comprehensive Chinese-language treatment appeared from Mu Yang on 36Kr on March 16 under the headline “How the AI Model Claude Entered the US Military’s Kill Chain.” The piece traced the full arc of Claude’s integration into CENTCOM, noting that US Central Command used Claude continuously throughout the operation for intelligence assessments, target identification, and battle scenario simulations — even though Trump had issued an executive order banning all federal agencies from using Anthropic’s tools the day before the strikes began. Internal Pentagon assessments, the piece reported, indicated that replacing Claude would take three to six months, and potentially up to twelve, given how deeply it had been customized for Palantir workflows and contractor software pipelines. Mu framed the episode as what he called “the Oppenheimer Moment in the militarization of artificial intelligence” — not a single detonation announcing a new era, but a threshold crossed through the deep binding of technology to the machinery of killing. “AI is not a tool of war,” Mu wrote. “It is becoming the form of war itself.”

Yu Zheng, writing in The Economic Observer, published a similar analysis on March 2, framing the Khamenei strike as a product of AI’s deep integration into the kill chain. The piece identified what it called a “civilian-to-military” inversion: unlike the Cold War, when military R&D produced technologies (the internet, GPS) that later found civilian applications, the current AI militarization is defined by commercial models trained in consumer markets being adopted wholesale by the Pentagon. Large language models built by OpenAI, Google DeepMind, and Anthropic, the article argued, now exceed the general intelligence of many specialized systems developed inside the military’s closed ecosystem — a dependency that has created what it termed an unprecedented “strategy-capital symbiosis” between Silicon Valley and the defense establishment.

Zhang Junse, a Chinese military expert interviewed by the Global Times on March 16, placed the use of AI in the Iran war on an evolutionary continuum stretching back through Gaza to the early days of the Russia-Ukraine conflict. In Ukraine, he argued, AI played a supporting role — helping with reconnaissance, target identification, and intelligence analysis, but with limited coordination between systems. By the time of the US-Israeli strikes on Iran, AI had fused with joint command-and-control architecture, enabling real-time data sharing and coordinated operations across land, sea, air, space, electromagnetic, and cyber domains simultaneously. His bottom line: the battlefield role of AI has shifted “from auxiliary to core, from local to global, and from passive to autonomous.”

The OFweek AI platform struck a more cautious note on March 2. It confirmed that Claude played what it called an "indispensable" role in CENTCOM's intelligence workflow — collating intelligence, identifying targets, and running battlefield simulations — but stressed that humans remained in the loop on final decisions throughout. The more striking observation was operational: the White House ordered a ban on Claude before the war began, but Claude was already so deeply woven into Pentagon systems that pulling it out mid-conflict was not a realistic option. Whether it can be extracted going forward, the piece noted, remains an open question.

How this impacts Chinese tech interests in the Middle East

Zhu Zhaoyi, director of the Middle East Institute at Peking University’s HSBC Business School, gave a wide-ranging interview to the Securities Times (published March 6 on Phoenix Financial) that addressed what the war means for Chinese corporate exposure in the Gulf. In his blunt assessment of the Gulf, he noted the UAE has been hardest hit among GCC states, for three reasons. Its deep normalization with Israel made it Iran’s highest-value target. Its economy depends heavily on foreign investor confidence. And it hosts a concentration of multinational regional headquarters whose presence rests on the now-damaged premise that the Emirates are a “safe oasis.”

Zhu argues that Chinese enterprises need to triage their interests in the Gulf. More specifically, he makes a case for separating between core assets worth defending and peripheral ones that can be pulled back. He predicted Chinese companies will pivot as a response to the war toward lighter-footprint engagement, likely focused on technology service contracts, management exports, and consulting agreements rather than direct equity or heavy infrastructure. He also expected a geographic redistribution, with investment migrating away from the UAE toward Saudi Arabia, Oman, and Qatar. None of this, in his view, signals a rupture of China’s commercial presence in the Gulf. The underlying logic of Chinese interests in the Gulf still holds. China needs energy and overseas markets, and the Gulf needs technology, infrastructure capacity, and renminbi diversification channels. But new projects, he said, will sit in a holding pattern until the security picture clarifies.

Gu Jiashi, an economics PhD writing on Guancha within 48 hours of the strikes (March 2), argued that the war's impact on the AI supply chain was the variable Chinese investors were underestimating. The Gulf, he wrote, provides three pillars to the global AI buildout: cheap energy to power data centers, sovereign wealth fund capital to finance them, and a geographic position at the data-transit crossroads linking Asia, Europe, and Africa. The AWS strikes demonstrated that all three pillars now sit inside a conflict zone. AI infrastructure has moved from bystander to battlefield target. And the cascading effects run in both directions — the Strait of Hormuz closure drives up energy costs for data centers globally, while Gulf sovereign funds facing reconstruction bills at home have less capital to deploy into overseas AI investments. The squeeze, in Gu's framing, hits compute from the supply side and the funding side at the same time.

Finally, a financial analysis from a Chinese netizen on East Money (March 7) made the investment case for China most explicitly. The AWS strikes, the author argued, highlight the core value of compute sovereignty. Domestically, Chinese government and enterprise users would accelerate migration off foreign public clouds and onto Alibaba Cloud, Huawei Cloud, and Tencent Cloud. Internationally, growing wariness of US chip controls and American cloud service geopolitical risk would create breakthrough opportunities for Chinese compute solutions — specifically naming Huawei Ascend clusters, domestically produced servers, and smart computing centers — in Southeast Asia, the Middle East, and Africa.

The Chinese commercial play is risky for Gulf states

The Chinese discourse on compute sovereignty reads the AWS strikes as a potenial market opening for Chinese enterprises, but the near-term value proposition for Gulf states to migrate AI workloads from Western clouds onto Chinese alternatives is weaker than Beijing’s analysts suggest. GCC governments are not in a position to execute a wholesale shift in cloud service providers in the middle of a kinetic conflict. The operational priority is continuity, not diversification. What the strikes have actually accelerated is a different kind of migration largely of data from Gulf-hosted data centers to offshore facilities in less exposed regions. Amazon itself advised customers to replicate workloads to other AWS regions outside the Middle East. The immediate logic for Gulf enterprises and government agencies runs toward geographic redundancy within existing provider ecosystems, not toward swapping ecosystems entirely. Onboarding a new hyperscaler, which includes rewriting APIs, recertifying security frameworks, retraining staff, and renegotiating data residency agreements, takes months under normal conditions. Under fire, this is not a serious option.

The sovereignty problem cuts deeper than the commercial one, and it runs against every available option. Gulf states — the UAE above all — built their AI strategies around a hard data sovereignty requirement: sensitive data stays onshore, processed on local infrastructure, under national jurisdiction. That principle drove the construction of domestic data center capacity in the first place. The war has blown a hole in the logic. Onshore infrastructure can be physically destroyed or taken offline by missile strikes, which means the sovereignty premise and the security premise now pull in opposite directions. Couture and Toupin argued in their 2019 taxonomy of digital sovereignty that sovereignty in the digital realm is not binary because it exists on a spectrum defined by where data is stored, who processes it, under whose legal jurisdiction, and who can compel disclosure. Using the UAE as a case, before this war, Emirati data on AWS servers in Abu Dhabi sat at the high-sovereignty end of that spectrum. When those workloads have to move offshore for survivability — whether to AWS regions in Frankfurt, Google facilities in Mumbai, or anywhere else outside the blast radius — the data slides down the spectrum involuntarily. The commercial relationship with the provider stays intact, but the legal and political control over the data does not. Abu Dhabi’s leverage over what happens to that data weakens the moment it crosses a border, and it does not matter whose cloud carries it. A Chinese hyperscaler hosting Emirati government workloads in Shanghai poses the same jurisdictional exposure as an American one hosting them in Virginia. The dependency rotates; it does not resolve.

The conflict has exposed that concentrating AI infrastructure within missile range of a hostile regional power both threatens the physical infrastructure and hardware, while also undermining the entire sovereignty architecture built on top of it. This creates a unique design problem GCC states: every option for protecting data and ensuring continuity of AI systems means trading one form of control for another. If they keep their data onshore, they retain jurisdiction but accept that their servers sit inside the blast radius. If they move it offshore, they gain survivability but surrender the jurisdictional control that sovereignty was designed to protect. If they distribute across multiple regions, they gain redundancy but fragment their data across competing legal regimes.

Subscribe now

The war in Iran has arrived at a particularly brutal moment for the Gulf’s AI ambitions. Gulf states were in the middle of a buildout exceeding $300 billion in data centers, chips, and AI infrastructure, backed by OpenAI, xAI, Microsoft, Amazon, Oracle, and Google. The Pax Silica initiative of January 2026 had just brought the UAE and Qatar formally into a US-led framework to keep advanced semiconductors away from China. The architecture seemed locked in.

Subscribe now

Then on March 1, Iranian drones struck three AWS data centers — two in the UAE and one in Bahrain. The IRGC claimed responsibility, framing the attacks as targeting facilities it said supported “the enemy’s military and intelligence activities.” Banking and payments failed, and enterprise software across the region went dark. AWS told affected clients to migrate ongoing workloads to other AWS regions and direct traffic away from the Middle East. It was the first time in history that data centers had been deliberately targeted as military objectives in an active conflict.

Iran has since escalated rhetorically. IRGC-affiliated Tasnim news agency published a list of “new targets” spanning 29 locations across Bahrain, Israel, Qatar, and the UAE, naming Amazon, Microsoft, Google, Oracle, Nvidia, IBM, and Palantir facilities explicitly as “enemy technology infrastructure.” The warning threatened that “...as the scope of the regional war expands to infrastructure war, the scope of Iran’s legitimate targets expands.” Included in the list are Nvidia’s largest R&D center outside the US in Haifa, Google’s Dubai regional office, IBM’s AI research hub in Be’er Sheva, Palantir’s Abu Dhabi collaboration center, Oracle’s Jerusalem, and Abu Dhabi offices, as well as more AWS facilities.

How the Tech Companies Are Responding

The hyperscalers have not said much publicly, but they have begun preparing for more potential attacks. Nvidia shut its Dubai office and shifted to remote work. CEO Jensen Huang sent a company-wide memo saying his crisis management team was “working around the clock and actively supporting affected employees and their families” in the Middle East. Amazon instructed all corporate employees in the Middle East to work remotely and “follow local government guidelines.” Dozens of Google employees were stranded in Dubai after the company’s cloud sales conference when commercial flights were cancelled, a company memo to staff called the situation “concerning.” Google’s public statement was careful to the point of saying very little: “The situation in the Middle East is evolving rapidly and we are monitoring it carefully. Our focus is on the safety and well-being of our employees in the region.”

It is safe to say that companies are now scenario planning if Iran acts on its threats. This could slow new capital deployments and pause some planned partnerships until it is clearer what comes next in the Iran conflict. Tess deBlanc-Knowles at the Atlantic Council told CNBC that rather than exiting the Gulf, companies are more likely to hedge. The sunk costs are too large for most to contemplate a full exit. But Patrick J. Murphy from Hilco Global argued that, instead, companies may accelerate projects in Northern Europe, India, or Southeast Asia, “where power supply, regulatory frameworks and security conditions are more predictable.” Geopolitical strategist Abishur Prakash told Rest of World: “This is all inverted now, exposing the entire technology landscape and ambitions of the region.”

The Security Architecture Quesiton

The strikes have exposed how datacenter security was largely focused on preventing ground-based incursions and external infiltration. It does not effectively factor kinetic air strikes. Ali Bakir, assistant professor of international affairs at Qatar University, told Rest of World: “The security frameworks underpinning the US-UAE AI partnership appear to have focused on supply chain control and geopolitical alignment, not on physical defense during high-intensity conflict.” Geopolitical strategist Abishur Prakash told the same outlet: “Strategic planning revolved almost entirely around energy and financial flows, leaving technology infrastructure vulnerable.”

The Pax Silica architecture was designed to keep Chinese chips out — not to protect data centers from ballistic missiles. Those are different problems requiring different solutions, and only one of them received serious policy attention before the war began.

The connectivity picture is equally exposed. Seventeen submarine cables pass through the Red Sea, carrying the majority of data traffic between Europe, Asia, and Africa. Additional cables run through the Strait of Hormuz. If these are cut off, as occurred following Iran attacks last year, then even the best functioning data centers cannot serve the larger customer base beyond the GCC, causing regional blackouts, not just in the Gulf.

Compounding the problem is the dual-use argument that Gulf data centers are legitimate military targets because they are being actively deployed for military purposes. Fortune reported that the US military used Anthropic’s Claude — running on AWS — for intelligence assessments, target identification, and battle simulations during the Iran strikes. If commercial cloud infrastructure carries military workloads, adversaries can and will treat it as military infrastructure — but international law scholars Klaudia Klonowska and Michael Schmitt argue in Just Security that Iran’s own stated rationale — striking the facilities “to identify the role of these centres in supporting the enemy’s military and intelligence activities” — may itself be unlawful, since the law of armed conflict requires that determination to be made before an attack, not by conducting one. That legal ambiguity offers cold comfort to the companies whose servers are burning. As Zachary Kallenborn, PhD researcher at King’s College London, told Fortune: “If data centers become critical hubs for transiting military information, we can expect them to be increasingly targeted by both cyber and physical attacks.”

What Comes Next

The structural question the war has forced into the open will not disappear when the shooting stops. Can the Gulf region build the world’s most ambitious AI infrastructure while sitting inside an active conflict zone, with no physical security framework covering those assets, and with a security architecture designed for a different threat entirely?

The short answer is yes. But there will be a moment of taking stock — and the Gulf is living through it right now. The immediate task is protecting the facilities Iran has already named, the staff inside them, and finding more effective ways to de-risk infrastructure that nobody designed for a shooting war.

It is also worth keeping the threat in proportion. Hitting a data center will, for now, do less damage than destroying oil and gas infrastructure. A single drone in a couple of AWS availability zones is not going to move the global cost of compute. Amazon has already demonstrated the countermeasure is viable by routing workloads to other regions, keeping outages temporary and local. A sustained blockade on Gulf energy exports does not have the same kind of redundancy and ripples through the global economy in ways a degraded cloud region simply does not. So yes, AI infrastructure will be targeted. Iran has made that plain. But the effect will be narrower than the other targets Iran is trying to hit, and the operators have more tools to absorb the blow.

Subscribe now

This is my first attempt of using Substack Recording Studio to create video contact. In this video, I unpack some of the challenges facing any aid response in Iran and some of the hurdles for scaling an aid delivery.

Subscribe now

Iran’s humanitarian crisis is growing quickly, with UNHCR estimating that as many as 3.2 million people may already be temporarily displaced inside the country. The needs are likely to be immense: shelter, food, water, fuel, cash assistance, trauma care, medicines, maternal and child health, sanitation, and support for host communities absorbing those fleeing the strikes. But the problem is not only the scale of need. It is that Iran is an extremely difficult environment for outside humanitarian response due to restrictions by the current authorities. Sanctions risk, banking restrictions, compliance burdens, and strict state controls over foreign funding sharply limit what can move and how fast it can move. The main viable delivery actor inside Iran is the Iranian Red Crescent, backed internationally by the IFRC, rather than a large U.N.-run operational presence.

At the same time, the regional logistics environment is deteriorating. The Gulf would normally be the most plausible platform for financing, staging, and moving aid, but Iran is actively striking the GCC, degrading the very Arab hubs that would usually support humanitarian delivery. That shifts attention to narrower and less efficient corridors, especially Turkey, Iraq, and Azerbaijan. Turkey in particular is already preparing for possible refugee inflows and treating the crisis as a humanitarian, border-management, and national-security challenge.

The hardest question, however, begins once aid enters Iran. How do aid providers track where it goes and verify that it reaches civilians in need when access for outside observers is limited? That is what makes this crisis so dangerous. The needs are rising faster than the available channels to meet it.

The preliminary emergency appeal is here:

*Please note that some may argue my comment on the Iranian Red Crescent being a state-run entity is incorrect (which I admit may have been the wrong framing), but the Iranian Parliament passed a law which requires the Chairperson of the organization to be appointed by the President of Iran.

Subscribe now

Get more from Jesse Marks in the Substack app

Available for iOS and Android

Get the app

Since March 9th, the commentary on Iran from Chinese commentators has exploded into a large number of topics including energy, Iran’s selection of a new leader, how mediation is going, the nuclear question, and so on. For this update, I am focusing on a smaller sample of views I found particularly interesting concerning mediation, Iran’s new leader, and changes in the Foreign Ministry’s increasing call for ending indiscriminate attacks in the Persian Gulf (without naming names). I conclude with how I think China’s mediation has met its low ceiling in the conflict and what it can achieve (and what it can’t).

Let’s dig in.

Subscribe now

Li Mingjiang, an associate professor at Nanyang Technological University’s S. Rajaratnam School of International Studies, argued in an interview with Lianhe Zaobao that Zhai Jun’s mission was primarily meant to stress “the hope of stopping the conflict as soon as possible and achieving a ceasefire.” He also observed that Beijing had offered Iran’s new leader “no obvious political support,” underscoring the caution of China’s response. Li’s interpretation was that China was trying to signal support for de-escalation without becoming politically overcommitted to the evolving leadership situation in Tehran.

Shen Dingli, a Shanghai-based international relations scholar interviewed in the same report, commented that Beijing is watching closely to see how Mojtaba Khamenei governs after succeeding his father. Shen’s core point was that Chinese observers are wary of whether Iran’s new leadership will continue the previous line of developing nuclear weapons “to confront Israel”. His piece carries a similar reading as Li’s, namely that China’s mediation posture is still a guarded one. Beijing will not simply mediate to defend Iran, but will first assess whether the new Iranian leadership itself is likely to follow the same ambition of creating a nuclear bomb.

Zhu Feng, dean of the School of International Studies at Nanjing University, argued also in an interview with Lianhe Zaobao that “in principle and morally, China is firmly on Iran’s side.” At the same time, he acknowledged the practical limits of Beijing’s diplomacy, saying Zhai Jun’s trip to the Middle East would likely not be able to stop continued U.S.-Israeli military strikes against Iran. Instead, Zhu said the mission could still serve a useful purpose by reducing misjudgments between Iran and Arab states and helping prevent the conflict from widening further around Iran. His view combined moral and political sympathy for Iran with a sober assessment of China’s limited leverage.

Long Chen, assistant researcher at the Chongyang Institute for Financial Studies at Renmin University of China, argued that the current conflict has since deviated from its predetermined course and is hurtling headlong into a direction no one can control. He comments that “…this war has never been a simple military standoff, but rather a power struggle, an entanglement of interests, and the displacement of countless civilians.”

Sun Degang, director of the Middle East Research Center at Fudan University, pointed out that the US plan in Iran was a swift victory, fearing that a prolonged war would affect the midterm elections. But when a swift victory is hopeless, a "dignified withdrawal" becomes the only rational option. The conflict is likely to transition to a "fight-while-talking" phase when both sides are exhausted, ultimately forming a temporary ceasefire framework acceptable to all parties.

Faisa, identified by People’s Daily Online as an expert for People’s Arabic edition, argues that China sees Zhai Jun’s current shuttle diplomacy as an effort to contain escalation in the Iran crisis, not end it. In her analysis (which closely mirrors state language), Beijing’s priority is to push all sides to halt fighting, prevent the conflict from widening across the region, and steer the parties back toward negotiations. Implicit in her writing is also a reputational angle, which she writes is to reinforce China’s image as a stabilizing power in the Middle East that favors political solutions, respect for sovereignty, and dialogue over confrontation.

Chinese FM Spokesperson Guo Jiakun responded on March 11 to an Iran question from a journalist that “China does not go along with attacks against countries in the Gulf region and condemns nondiscriminatory attacks against civilians or non-military targets. The pressing priority now is to stop military operations at once and prevent the conflict from spreading. The way out of the conflict is to return to dialogue and negotiation as soon as possible and strive for restoration of peace. The fundamental solution is to jointly return to the right track of abiding by international law and basic norms of international relations.”

He goes on to describe the current mediation: “Since the very first day of the conflict, China has unequivocally called for ceasefire, end of hostilities, return to dialogue and negotiation, and political settlement. As we speak, Special Envoy of the Chinese Government on the Middle East Issue is in the Middle East to carry out shuttle diplomacy. As a permanent member of the UN Security Council and sincere friend of Middle Eastern countries, China will stay committed to urging peace and defending justice and fairness. China will continue to strengthen communication with relevant parties, including parties to the conflict, and play a constructive role for deescalation and restoration of peace.

My takeaways over the past few days

I argued in a 2024 bookchapter on Chinese “strategic facilitation” in the Persian Gulf:

“China, however, has yet to provide any meaningful guarantees to offset the concerns of Tehran and Riyadh, particularly in the realm of security. One of the primary miscalculations in China’s perspective on the Gulf crisis is the extremity of the security dilemma at play in the Gulf. Among the primary drivers of the Iran-Saudi rivalry, military competition, which has proliferated into a regional competition for influence among neighboring states and nonstate actors, continues to be one of the major impediments to resolution.”

This has been the ceiling of China’s diplomacy for nearly a decade. When an Iran-linked drone slammed into Saudi Aramco oil infrastructure, China avoided assigning blame or aligning itself openly with either Tehran or Riyadh. Instead, it pursued a calibrated middle course: urging restraint and de-escalation while muting U.S. and Saudi accusations of Iranian responsibility. As one analyst put it, Beijing was “unwilling to become embroiled in the rivalry,” preferring to deepen ties with both sides in order to project deliberate neutrality.

Seven years later, Iran, in retaliation for U.S. and Israeli strikes, is raining missiles and drones across the GCC, targeting not only oil infrastructure but also civilian buildings and critical infrastructure. In the present conflict, there is no viable or plausible way for China to frame Iran’s attacks on the Gulf without assigning some degree of blame to Tehran. To be sure, the United States and Israel initiated this specific round of fighting, and Chinese scholars generally agree that China will not attempt to mediate directly between Washington and Tehran. The outer limit of its diplomacy, for now, is to call on all parties to return to the table. But Chinese authorities are approaching the limits of neutrality when it comes to Iran and the GCC. But, any such credible condemnations of Iran will run up against the redline of other Chinese authorities–lending any any credibility to U.S.-Israeli strikes.

The evidence is already visible in the Foreign Ministry’s public statements. When Guo Jiakun states that “China does not go along with attacks against countries in the Gulf region and condemns nondiscriminatory attacks against civilians or non-military targets,” he is indirectly referring to Iran in addition to the United States and Israel—not the GCC. The remarkable restraint shown by Gulf states has resonated in China, as it has elsewhere. Many would argue that such clear violations of national sovereignty, especially direct attacks by Iran, would provide reasonable grounds for GCC retaliation. But the region understands the escalatory spiral that such a response could unleash, and so does China. Beijing may face pressure to be louder, not quieter, in condemning Iranian actions. Doing so will unavoidably mean placing Iran alongside Israel and the United States, something China has thus far refrained from doing. Beijing is reaching the low ceiling of its mediation role.

So, the big question: Could China be the broker of Middle East peace? I personally remain skeptical. The heavy lifting of mediation and diplomatic outreach is still led by the GCC, even in the face of Iranian rocket, missile, and drone attacks. One of my biggest takeaways is that, even in the midst of this war, the GCC mediation narrative that many concluded the Gulf had adopted as rhetoric or cheap PR wins has proven resilient in the face of major escalation. If this war ends through diplomatic means, it will likely be the Gulf Arab states who broker it.

Finally, where is Chinese mediation going to be most applicable then? I suspect we will see more intensive Chinese mediation occuring on the periphery of the Iran conflict to prevent the spread of conflict eastward into Pakistan and Afghanistan, where flareups are already underway. There are reports that China is stepping in to help reduce tensions between the two neighbors and prevent a spillover. This will be an area to keep watching.

Thanks for reading. Please like, comment, share, subscribe, and, consider supporting more of my work.

Forward: I have been very vocal about China needing to play a large role in mediating in the Persian Gulf given its most important regional economic partners (the GCC) are actively being attacked by its strategic partner of choice (Iran) in the region, who is being attacked by their most important global competitor (the US). There are a number of competing views which quesiton the extent to which China has the will and leverage to do so. I have written extensively on Chinese mediation in the Persian Gulf and will be the first to highlight those limitations, both in terms of capacity and interest. Yet, despite these challenges, Zhai Jun, China’s Middle East Special Envoy, is in the region doing mediation but it is not clear exactly what that means yet. It does not mean China mediating between the U.S. and Iran, but rather China helping end Iran’s assault on the GCC.

The Zhai Jun mission is the first physical deployment of a Chinese envoy to the conflict zone since Operation Epic Fury began. It is a major step up from telephone diplomacy signaling that Beijing sees the situation is grave enough(or they see enough opportunity) to require a direct regional presence. For now, that is the best China can deliver to the GCC and Iran.

What Chinese media has noted

Spokesperson Mao Ning announced at the March 5 Foreign Ministry press conference that China would send Jun to the region in the near term to make active efforts toward de-escalation. Speaking with his Saudi and UAE counterparts on March 4, Wang Yi stated that China remained a force for peace and was willing to continue playing a constructive role, reiterating that war spillover served no party’s interests and that civilian protection, energy infrastructure safety, and freedom of navigation were all non-negotiable red lines. Wang specifically praised the GCC’s emergency foreign ministers’ meeting for affirming that dialogue and diplomacy were the only paths through the crisis.

Li Zhenjie, associate researcher at the Mediterranean Studies Institute of Zhejiang International Studies University, told Sputnik Chinese on March 4 that China occupies a structurally unique position in the crisis precisely because Washington has disqualified itself as a mediator. With the United States as the conflict’s provacator and regional powers too internally divided to play a bridging role, China can project itself as a responsible major power capable of maintaining effective communication with all Middle Eastern parties and the major European states, giving it a rare platform for crisis management. He argued that Beijing’s public reaffirmation of support for Iran’s sovereignty following Khamenei’s death served a concrete stabilizing function. He argued It helped shore up morale within Iran’s transitional leadership and demonstrated to Tehran that it was not internationally isolated. Li also framed China’s mediation posture as speaking beyond the bilateral relationship. From his view, Beijing’s calls to stop the fighting amounted to an appeal on behalf of every country touched by the spillover (the GCC and other Arab states).

Mapping the Opportunities and the Limits of Mediation

China is not sending Zhai Jun to stop the war. No Chinese envoy is going to walk into Riyadh or Abu Dhabi and persuade Washington to stand down its carrier strike groups or convince Jerusalem to halt a campaign it has spent years planning. Beijing knows this. The core questions we have to answer: 1) What can Zhai Jun credibly achieve within these constraints? 2) Is the mission calibrated to those realistic objectives or to something more ambitious? If the latter, failure could expose the limits of Chinese influence rather than demonstrate it.

There is the potential for reputational gain through mediation, but the path is narrow. In the moment of writing this, China is the only major power with working relationships on all sides of this conflict. Washington has, at least for now, disqualified itself as a mediator by becoming a combatant. Europe lacks the regional standing and the political will; its response to Epic Fury has been a combination of strongly worded statements and institutional paralysis. Russia’s deep ties to Iran compromise whatever credibility it might otherwise bring to the Gulf states, who are simultaneously absorbing Iranian strikes and watching Moscow say nothing about them. That structural vacancy is real, and Beijing’s decision to fill it — keeping lines open while everyone else either joined the fight or receded into irrelevance — cannot be easily dismissed. When we divorce analysis from the western frame of reference and look through the lens of the Persian Gulf states, China’s neutrality, while sometimes frustrating for each side when the other escalates, has broadly been accepted. As I have previously argued, China’s strategic balancing in the Persian Gulf has taken years to build and could be undermined in a matter of weeks if not carefully protected. With this in mind, Zhai Jun has to walk a tightrope, but the pressure to deliver something for Chinese national interests is growing.

There is also a Global South dimension Beijing is clearly playing to. After Venezuela and now Iran, the narrative of US militarism in the international community is writing itself. Two governments in two hemispheres, both adversarial to Washington, both targeted by American military action within months of each other — the through-line is not subtle, and it lands. China presenting itself in a mediator role aims to feed directly into that story, projecting Beijing once again as a willing, neutral third party between Iran and the GCC states. How much that narrative actually moves regional governments when rockets are flying, shifts the region’s changing alignments, or produces measurable diplomatic outcomes is genuinely hard to quantify. Reputational gains of this kind are diffuse, slow-moving, and rarely decisive in the near term. But they compound. Beijing, despite its uncertainty about how to navigate its current position, is playing a longer game with the Gulf region than US leadership seems capable of at this moment in the conflict.

That does not remove the risks of attempting mediation. Showing up creates expectations, and expectations that go unmet could carry their own costs. If Zhai Jun returns without something Beijing can point to - the gold standard would be a permanent reduction in Iranian strikes on Gulf infrastructure or even a joint statement that signals forward movement - the mission becomes its own liability. China risks looking less like a capable stakeholder and more like an ineffectual bystander. Whether that actually damages Chinese standing with Riyadh and Abu Dhabi in any serious way is unclear.

For China, showing up as a mediator (even if it purely rhetorical) was already half the game. No other major power with Beijing’s relationships and clout across the region has been willing to try. But showing up is where the easy part ends, and being clear-eyed about what China can actually accomplish — and what it cannot — is the only honest basis for judging whether the mission succeeds or fails.

The structural constraints are severe. There is a US armada in the region near the Strait of Hormuz. It is actively bombing Iran. Iran wants to sink it. China has no ability to change any of those facts, and no diplomatic instrument that bears on the military balance between Washington and Tehran. Beijing cannot stop the strikes. It likely cannot guarantee safe passage through a strait that two armed forces are actively contesting (though they may try). It cannot negotiate a ceasefire that the US and Israel have not already decided to accept. These are the hard ceiling of what Chinese diplomacy can reach in a conflict where the primary belligerents are not listening to Beijing and have no particular reason to start.

The more concrete and achievable gain is on the Gulf side of the equation. Beijing does not need to stop US and Israeli strikes to protect its core interests. It needs to prevent the conflict from consuming the Gulf economic architecture it has spent a decade investing in. Beijing may actually be able to do something about that. There are real limits to what that translates to in terms of tangible support for the Gulf — China will not be providing interceptors to shoot down Iranian drones — but it could convince Iran to stop firing them.

If Zhai Jun can credibly communicate to Tehran — through whatever channels exist — that continued strikes on GCC infrastructure risk the one major power relationship Iran still has, that carries some weight. Iran needs China more than China needs Iran, and Beijing has rarely been willing to use that asymmetry directly. A genuine signal that GCC strikes are a red line for Beijing would be the single most concrete outcome the mission could produce. Iran’s leadership appears to have agreed to stop targeting the GCC in the past 48 hours — but it is too early to attribute that to Chinese mediation. My early read is that it may be one driver, but the larger factor is the GCC’s own insistence that it played no role in US strikes on Iran, combined with continuous pressure on Tehran to stand down before Gulf states are forced into the conflict directly.

The mediation mission succeeds if it preserves Chinese access to all parties, produces at least one visible diplomatic output Beijing can point to, keeps GCC confidence in Chinese partnership intact, and signals clearly enough to Tehran that strikes on Gulf soil carry a cost in the relationship that actually matters to Iranian survival. That is a modest definition of success.

Subscribe now

In Kenneth Waltz’s levels of analysis, international relations students learn early to examine international relations at three levels: the global international order (the system), the states which comprise it (the state), and the substate organs of power (the man) — leaders, institutions, voters, civil society, the military. These all shape how states choose to interact and, in some cases, go to war. Of these, the one we fail to spend sufficient time exploring is the psychology of the leader making decisions, the circle of influence around that leader, and the dynamics which shape their thinking such as their world view, core assumptions, decision-making frameworks, and degree of rationality and logic applied to both concrete and abstract decisions. In practical politics, readers who work in or near government understand these dynamics because passing policy is often an art of reading a leader, his advisors, and the structures which frame their worldview. I want to lean in on this point by drawing on the work of the late Aaron Rapport, a great international relations scholar (and my former professor) who passed away in 2019 after a long battle with cancer, and whose contributions to political science had an outsized role in my own analytical process.

Subscribe now

Rapport’s signature contribution was the application of construal level theory — CLT — to strategic assessment. His argument, tested against the Bush administration’s preparations for postwar Iraq, ran as follows: when leaders conceive of military campaigns as sequential phases, they evaluate near-term operations concretely, focused on how things will be executed, attentive to feasibility, receptive to critical information. But they evaluate distant operations abstractly, focused on why they are fighting, fixated on the desirability of their goals, reliant on preexisting beliefs, and resistant to concrete warnings about costs.

Three mechanisms drive this pattern. First, abstract construal privileges desirability over feasibility: leaders can articulate what they want from the endgame but struggle to articulate how they will achieve it. Second, communication fluency shifts: leaders in an abstract mindset become unreceptive to concrete information challenging their postwar assumptions. Third, leaders rely on simplified mental models drawn from prior experience rather than case-specific evidence, which obscures how a given situation departs from the template in their heads.

The parallels between Iraq 2003 and Iran 2026 are striking. In some ways the current case is more extreme.

Start with the sequential framing. President Trump has explicitly projected a combat timeline of “four to five weeks” and framed the operation around defined military targets: the navy, the missile program, nuclear infrastructure, IRGC leadership. The “day after” is not merely neglected — it is actively pushed into a separate cognitive category. When Defense Secretary Hegseth told the Pentagon press corps on March 3 that “this is not Iraq, this is not endless — this operation is a clear, devastating, decisive mission,” he was performing exactly the framing operation Rapport identified as the precondition for abstract construal of later phases. The sharper the separation between “combat” and “what comes after,” the more abstract the leader’s assessment of the latter becomes.

Now consider the first mechanism: desirability over feasibility. Trump can articulate his goals with confidence. Destroy the missile program. Eliminate the nuclear threat. Sink the navy. Give Iranians their “freedom.” But when asked on March 4 about the worst-case outcome, the president replied: “I guess the worst case would be we do this and somebody takes over who’s as bad as the previous person, right? That could happen.” He then acknowledged that “most of the people we had in mind” as potential successors “are dead,” adding, “Now we have another group. They may be dead also, based on reports. Pretty soon, we’re not going to know anybody.” These are not the statements of a leader who has assessed feasibility. They are the statements of a leader so captured by the desirability of his goals that the gap between ends and means has become invisible.

The communication fluency problem — Rapport’s second mechanism — is equally acute. Congressional briefings have produced bipartisan fury. Senator Mark Warner observed that the war’s stated goals have shifted four or five times. Senator Blumenthal left the classified briefing “more fearful than ever.” House Democrats, less charitably, called the briefing “bullshit.” But these concrete warnings are not landing. The administration’s cognitive frame filters them out. This mirrors what Rapport documented with Jay Garner’s pre-invasion Phase IV (a term used to describe post-conflict operations) briefing in 2003, at which President Bush and NSC principals asked no questions and struck the head of postwar planning as uninterested in his mission.

The third mechanism — reliance on simplified beliefs — manifests as the Venezuela template. Trump appears to be mapping his experience with the Maduro extraction onto Iran: decapitate the regime and a cooperative successor emerges to do business. This simplified mental model obscures how radically Iran’s political system — with its layered clerical-military hierarchy, its eighty-eight-member Assembly of Experts, and a deeply institutionalized Revolutionary Guard — departs from Venezuela’s autocratic structure and the extent to which Iran has prepared for the current scenario. The Israeli strike on the Assembly of Experts while it was selecting a new supreme leader suggests the coalition is improvising the destruction of the succession mechanism without a theory of what replaces it. Rapport would recognize the pattern instantly: a leader deducing outcomes from a general belief about regime change rather than from the concrete specifics of the Iranian case or any empirically-informed understanding of what comes next.

There is one dimension where the Trump case diverges instructively from Rapport’s Iraq analysis. Bush and Rumsfeld had arguably longer time horizons than the Trump administration when they undertook the Iraq war. This is not become they were planning effectively, but generally, they sought democratic transformation of the Middle East and long-term deterrence. Their abstract construal of the endgame followed from the depth of their investment in transformative goals. Trump’s time horizons oscillate. His stated objectives swing between transformative ambition — “freedom for the people,” regime change — and narrow military targets. Hegseth insists “this is not a so-called regime change war,” then adds, “but the regime sure did change, and the world is better off for it.” This incoherence may represent multiple cognitive frames competing within a single decision-making unit, with Rubio and Hegseth articulating modest military objectives while Trump and allies like Lindsey Graham voice transformative aspirations. The problem is that the operation’s design — killing the supreme leader, bombing the succession body — is a regime-change operation regardless of what anyone calls it. The mismatch between stated objectives and operational reality is itself an artifact of the construal level theory. In other words, the abstract framing produces actions that presuppose transformation while the concrete justifications trail far behind.

Rapport wrote that leaders who believe postwar conditions will be favorable tend to outsource Phase IV (post-conflict stabilization operations) to the population they have just bombed. Rumsfeld expected Iraqi exiles to rapidly establish authority. The administration assumed that the Iraqi army would help the coalition keep order. Trump’s February 28 video to the Iranian people — “When we are finished, take over your government” — is the same style move, but this time applied to a country of eighty-eight million people whose entire leadership structure is being systematically destroyed from the air. And, worst yet, if reports true, the Trump Administration is using the CIA to arm Iranian-Kurdish groups in Northern Iraq to open up a ground front in the war.

Aaron Rapport gives us a helpful framework that explains, with uncomfortable precision, why the United States keeps repeating the same mistake as before. There is extraordinary confidence by the U.S. leadership about the kinetic phase of war paired with almost no articulation of how political order reconstitutes itself after the smoke clears. Five days into Operation Epic Fury, we are watching that framework play out in real time.

Subscribe now

Discourse among Chinese scholars has expanded since the war began to focus on energy security and the question of Hormuz as well as how the U.S., Israel, and Iran are fighting the war. I have summarized a longer list of comments, quotes, and opeds (some in English, some in Mandarin) from a wider range of Chinese voices in Chinese media and, more recently, in Russian media.

Subscribe now

The “law of the jungle”

Henry Huiyao Wang, Founder and President, Center for China and Globalization (CCG and former Counselor to China’s State Council went on Bloomberg to comment on the strikes. Wang declared that “Now, we’re back to the law of the jungle.” and described the killing of Khamenei as opening “Pandora’s box for whatever we’re going to see in the future, because all countries fear this unilateral approach.” On China’s potential response, Wang warned that Beijing “won’t just accept it voluntarily” if its interests are deeply harmed, noting China “has many other means” and “has always been able to counter with sanctions, retaliate, and fight back.” He cautioned that a prolonged conflict could produce “a butterfly effect that plunges the entire world into recession.” Wang also argued the crisis makes the upcoming Trump-Xi summit more urgent, not less, pointing to China’s role brokering the Iran-Saudi Arabia deal as evidence Beijing has “deep interests in that part of the world.”

Chen Qi, Director of the Tsinghua Center for China-US Relations commented that the US action “indicates the failure of the third round of negotiations with Tehran to reach a basic consensus.” He argued Washington’s primary goal of the strikes are pressuring Iran into future concessions “while also diverting attention from domestic political issues in the United States, including the Epstein scandal, and building momentum ahead of the upcoming midterm elections.” He argued this “opened Pandora’s box” and further intensification of the crisis will “hinge on Iran’s capacity to retaliate.”

Lü Xiang, Research Fellow at Chinese Academy of Social Sciences (CASS) assessed that Washington’s regime-change objective “is unlikely to be achieved quickly,” noting that “Iran had long ago put in place mechanisms for a rapid leadership transition.” Lü emphasized that “any effort to cripple Iran’s military capabilities would require a prolonged campaign, not a short operation.”

Zhou Dewu (周德武), former deputy editor-in-chief of Ta Kung Pao, argued that wars are “rarely ended by the initiator”; rather, they end when the side under attack accepts defeat. He warned that if Iran’s new leadership rejects reconciliation and instead turns to asymmetric retaliation — including possible terrorist tactics — the region could face a renewed wave of extremism. He also described the targeted killing of foreign leaders as “a dangerous precedent in international relations,” one that creates a “chilling effect” in global politics and could “accelerate an arms race.” In his view, the strikes also deepened the global “trust deficit” around negotiations. “This is not only Iran’s tragedy,” he concluded, “but also a tragedy for the international community.”

Iran’s regime resilience and China’s strategic calculus

Gao Zhikai — Chair Professor at Soochow University and deputy director of the Center for China and Globalization — delivered perhaps the most forceful condemnation of the strikes in an extensive Guancha interview published on March 3. He described the attacks as “赤裸裸的侵略行径” (“naked aggression”) and the assassination of Khamenei as “国家恐怖主义” (“state terrorism”). Rejecting the logic of regime change, Gao argued that U.S. policymakers “mistakenly believe that removing Khamenei would cause Iran to collapse,” when in fact “Iran is a civilization of five or six thousand years and a pillar of Islam. These American ideas are completely divorced from reality.” He also warned of nuclear escalation, noting that without the late supreme leader’s restraining fatwa, a new leadership might conclude: “If you want to kill me, if you want to destroy this country, this nation, this culture, then I will fight you to the bitter end.” [你要杀我，你要毁掉这个国家、民族、文化，那我就跟你拼到底。]

Qian Yaxu, a researcher at the Charhar Institute (察哈尔学会) and Southwest Jiaotong University’s Center for Regional and Country Studies, argued in an interview with Sputnik Chinese that Iran is “a key node country on the western route of China’s Belt and Road Initiative,” and that instability there would create “enormous resistance” for the BRI. On the question of regime change, she assessed that Iran is “most likely” to avoid such an outcome because Khamenei had already prepared a succession plan. She added that if regime change did occur, “the US would maximize its interests in the Middle East,” allowing Washington to redirect more strategic resources and attention to the Asia-Pacific and intensify its containment of China. In her view, the crisis will also “inevitably impact China-Russia relations” by deepening strategic coordination between Beijing and Moscow.

Military and defense analysis

Wang Yanan, editor-in-chief of Aerospace Knowledge magazine in Beijing, offered a detailed assessment of Iran’s military options. On Tehran’s ability to disrupt the Strait of Hormuz, he argued that “even conventional anti-ship missiles and drones could pose a credible threat,” and that enforcing a blockade “would not necessarily require highly sophisticated means.” He described Iran’s air defenses as being organized in a “mosaic” structure that operates in a manner similar to guerrilla warfare, meaning that even if the capital or certain facilities were struck, its military command system and key equipment could continue functioning. In strategic terms, Wang suggested that Iran’s best course would be to demonstrate both resilience under attack and the capacity to sustain retaliatory operations over time, thereby frustrating the “blitz-style” political objectives envisioned by Washington.

Lan Shunzheng (兰顺正) — a Chinese military expert — told Sputnik Chinese on March 3 that “Iran absolutely possesses retaliatory capability,” including ballistic missiles, suicide drones, and reconnaissance-strike UAVs that “could pose a substantive threat to the US and Israel,” and could even make them “pay a heavy price.” He assessed that Iran’s current strategy is to “inflict maximum American casualties to spark domestic anti-war sentiment, thereby compelling the US to halt operations.” At the same time, he cautioned that Iran’s stockpile of advanced weapons is limited and “may be rapidly exhausted after several days of intense combat,” potentially producing a “断崖式” (“cliff-like”) drop in capability.

Energy security and economic impact

Wan Zhe — a professor of economics at Beijing Normal University — argued that the logic of international oil prices has shifted “from being driven by supply and demand to being driven by geopolitics,” with the “geopolitical premium” now serving as the core driver of short-term, pulse-like price increases. She also noted that prolonged instability in the Middle East could push some countries to slow more aggressive coal and oil phaseout policies, even if higher oil prices improve the economic attractiveness of solar, wind, and electric vehicles.

Wang Lei, an assistant researcher at the Institute of World Economics and Politics at CASS, described a Hormuz blockade as a high-leverage and highly self-destructive option for Iran. He notes that such a move would simultaneously disrupt major energy buyers dependent on the strait, harm neighboring states, and likely provoke stronger military escort operations and counter-blockade measures.

Dong Shaopeng, a senior research fellow at the Chongyang Institute for Financial Studies at Renmin University of China, told CNBC that U.S. actions in Iran had further eroded trust between Washington and Beijing. Although he still expected a Trump–Xi meeting to proceed, he expressed hope that the conflict would not spread to other countries in the Middle East.

The use of AI in the Iran war

One area I found particularly interesting was the unfolding conversation about the U.S.-Israeli use of AI in the ongoing Iran conflict for targeting. Responding to these reports, Liu Wei, director of the Human-Machine Interaction and Cognitive Engineering Laboratory at Beijing University of Posts and Telecommunications, told Global Times on March 3 that while AI can assist military operations and improve efficiency, “the decisive factor should always be human beings.” The article is explicitly framed around reported U.S.-Israeli AI use in operations against Iran, including claims that Anthropic’s Claude was used through Palantir-linked systems to help shorten the “kill chain,” and that Israeli AI use in Gaza formed part of the wider backdrop to the debate. Liu argued that AI’s strengths lie in processing massive data, high-speed computing, target recognition, and supporting sustained unmanned operations, but warned that it must remain firmly under human control. In his view, current AI’s “data-driven rationality” cannot reliably distinguish battlefield deception, ethical gray areas, or shifts in public sentiment; without “human intuition to see through enemy disguises, flexible strategy to set the limits of strikes, and moral principles to restrain lethal impulses,” AI risks becoming “a blunt weapon harming both sides.”

Xiang Ligang, director-general of the Zhongguancun Modern Information Consumer Application Industry Technology Alliance, in the same article argued that “the underlying logic of future warfare is undergoing profound change.” He points to the Persian Gulf, as well as examples from Ukraine, India-Pakistan clashes, and Gaza. Xiang said AI is now deeply embedded in modern conflict. He identified seven defining characteristics of future warfare — systematization, modularization, intelligentization, miniaturization, precision, unmanned operations, and low cost — with precision at the center, arguing that advances in AI-enabled positioning, communication, sensing, and identification are making precision strikes increasingly feasible.

My concluding thoughts

China’s discourse on the Iran conflict has moved in two directions: energy security and observing warfare tactics. On the former, China’s concern over energy will deepen as the war goes on. This is a major strategic vulnerability. Beijing is currently looking at both risk exposure and how to manage it, especially finding ways to limit the war’s energy impact on both China’s economy and the global economy. We will likely see China explore whether it can access Saudi Red Sea crude from Yanbu as a partial hedge if disruption through Hormuz worsens. At the same time, it is likely Beijing will be more vocal about keeping the straits of Hormuz open–this is one of the few shared U.S.-Chinese interests.

On the latter, Chinese observers are carefully watching how the war is being fought as a live demonstration of U.S. tactics. I have found fascinating play-by-plays by Chinese netizens trying to follow and interpret U.S. actions and signals from the region. Particularly revealing was the attention some Chinese technology analysts gave to reports of U.S.-Israeli use of AI in target generation—an issue I will return to in a future article. The U.S. has long warned about China’s AI ambitions and the possible military application of those capabilities on the battlefield. Yet Chinese observers appear to be wrestling with many of the same concerns, especially over the troubling precedents the U.S. and Israeli deployment of such systems may set for the future of warfare. Chinese strategists have long studied U.S. military operations, and lessons drawn from this war will shape how they think about conflict far beyond the Gulf.

Subscribe now

A quick note: this analysis aims to inform ongoing research and discourse during the ongoing war over how a wide range of actors view the ongoing conflict and bringing that conversation into English. It does not serve as a definitive assessment, policy prescription, or endorsement of any actor involved.

Editors Note: There was a mix up with some of the URLs (and my translation), which has been corrected and edits made. Thanks for those who flagged these issues.

I have received a large volume of questions on China’s view on the Iran war. It is diverse. I figured that in addition to ’s great piece on this (go subscribe to his substack), I would also add a review of the Chinese experts I am following as well as my own concluding thoughts and observations.

Subscribe now

When the United States and Israel launched Operation Epic Fury on February 28, killing Supreme Leader Khamenei and forty senior Iranian officials in a single night, China’s foreign ministry condemned the strikes within hours as a “flagrant violation of international law.”

The rhetoric was predictable. What was less predictable is what China’s scholars, military analysts, and state media commentators said next for a domestic audience. They asked the harder question: does Beijing actually need the Islamic Republic to survive?

Their answer? Not necessarily.

The dialogue emerging from China’s strategic community is uncomfortable for those who assume Beijing and Tehran stand in ideological solidarity.

The Consensus View — and Its Limits

A dominant Chinese analytical position holds that the Islamic Republic will not collapse. Tang Zhichao of the Chinese Academy of Social Sciences declared on China News Network that the negotiations preceding the strikes were “completely a smokescreen, just like in June last year,” identifying US regime change as the explicit objective. Fan Hongda of Shaoxing University, one of China’s foremost Iran specialist, called the changes of regime change through the Ayatollah’s assaination “nearly impossible”, arguing Iran’s political foundations are too deeply rooted in religious, social, and historical structures for external military force to uproot.

Liu Zhongmin of Shanghai International Studies University assessed that short-term regime collapse is unlikely without a ground invasion — but added a caveat that should concern Washington and Beijing alike. The crisis may trigger “factional splits, power grabs, or political restructuring” representing “a long-term hidden danger.” He goes on to argue, “The advantage of a ‘quick strike and quick retreat’ is avoiding being dragged into war, but if the goal is to transform Iran, the issue becomes more complex. Is it to preserve a "backboneless" Islamic Republic of Iran, softened towards the West? Or is it to completely change its political system and create a completely new Iran?”

He expresses doubt that Trump’s approach can achieve true regime change. The regime may survive the strikes. The system may not survive the regime.

This is the analytical gap Chinese scholars are beginning to fill.

The Economic Case for Flexibility

Qin Tian of CICIR noted that the strikes combined “military means and color revolution methods,” signaling Washington intends to shape whatever successor configuration emerges. The question Qin raises is one that Beijing will have to eventually answer, whether that successor threatens Chinese interests more than the Islamic Republic protects them.

The answer from China’s financial commentors is also revealing. Writing on Huxiu, analyst Xiao Xiaopao pointed out that Iranian oil accounts for only roughly 13.4% of China’s seaborne crude imports, purchased primarily by small “teapot” refineries in Shandong, not major state enterprises. China has real interests in Iran, she argued, but they are not vital. Singapore-based columnist Yu Zeyuan put the logic more plainly: “Regardless of who is in power, it is unlikely economic ties with China would be completely severed. This is the main reason China’s official response has been relatively calm.”

A more open Iranian economy, whatever government emerges, still points toward China because geography, infrastructure dependencies, and market scale make Chinese partnership the path of least resistance. From this view, the Islamic Republic is useful, but it is not irreplaceable.

The Hawkish Strand in Beijing

Most striking is how publicly some Chinese analysts have questioned Iran’s value as a partner altogether. Zhan Hao, a nationalist commentator on Guancha, argued Tehran had systematically failed to honor the 2021 25-year comprehensive cooperation agreement, preferred Russian weapons, used the agreement as leverage with the West, and leased Persian Gulf islands to India rather than China. His conclusion was blunt: the Middle East is “an important strategic interest area, but absolutely not a core interest.”

Tian Wenlin of Guancha reinforced the pattern after the 12 day war arguing that after Iran reached the 2015 JCPOA with Washington, “Iran’s attitude toward China immediately cooled.” The implication is pointed — any Iranian government’s orientation toward Beijing depends not on ideology but on external pressure. Tehran turns to China when it has nowhere else to turn. Sanctions are China’s greatest source of leverage in the relationship. Sanctions relief, paradoxically, dilutes it.

The Precedent Fear

If Chinese analysts are privately sanguine about who governs Iran, they are genuinely alarmed about what the operation signals for great-power competition more broadly. Cui Shoujun of Renmin University stated flatly that China would not provide military or security support — but identified the deeper anxiety: “China certainly has this concern — that the US will target nations it deems hostile with surgical strikes to seek regime change” and that this could “become a routine tactic.” Ding Long of Shanghai International Studies University warned the US and Israel would employ “salami-slicing tactics to continuously weaken the state, incite domestic unrest and jeopardize Iran’s security” — a template Chinese strategists recognize as applicable well beyond Tehran.

Shen Yi of Fudan University framed the operation with the Chinese idiom “smashing the cup as a signal” — a prearranged move disguised as diplomacy. He then posed the question Beijing is privately turning over: “Will large-scale internal change occur within Iran?”

The view from the military

PLA-affiliated and military analysts have focused on operational and strategic implications rather than Iranian domestic politics. Zhang Junshe, a retired PLA Navy senior captain and military affairs commentator told the Global Times that “By attacking the Iranian Navy, the US likely aims to completely take control of the Gulf region.” He assessed a ground war as unlikely — “the US is unlikely to get bogged down like in Iraq and Afghanistan” — and predicted Iran would “deploy missiles and UAVs and may mobilize Axis of Resistance forces.”

Wang Yunfei, a military affairs expert analyzed the intelligence dimensions on March 1: “Iran was said to have been heavily infiltrated, leaving little room for secrecy, especially regarding senior leaders’ movements.” He assessed that while US-Israeli forces “far surpass Iran militarily and technologically,” Iran’s retaliatory capabilities “exceeded expectations,” particularly the Fattah hypersonic missile with an interception success rate “seemingly below 30 percent.”

What Beijing Wants from the Rubble

Li Shaoxian of Ningxia University connected the Iran strikes to Trump’s January success in Venezuela — two regime-change operations in two months, each combining military and economic coercion. Fan Hongda’s framework for Washington’s real objective is the most analytically useful on offer from Chinese scholarship: not an overthrown Iran, but a “spineless” one — a regime making major nuclear concessions while remaining nominally intact, selected from among Iran’s surviving factions. China can work with that outcome. It prefers a strong Iranian counterweight to US regional dominance, but it has thrived inside a sanctions-constrained Iran for a decade.

Sun Degang of Fudan University and Pan Guang, Senior Advisor to the Chinese Association of Middle East Studies, have looked at how this is shaping of regional order over regime in Iran. Sun warned that Tehran might abandon "limited war" because this engagement concerns the "survival of the regime," potentially blockading the Strait of Hormuz and disrupting global supply chains. Pan argued the strikes constitute "a severe shock to the post-World War II international order."

In this view, the collapse of regional order and Gulf chaos— blockaded Strait of Hormuz, disrupted supply chains, cascading instability across the Gulf states where China has invested heavily in Vision 2030 integration—threatens China's broader Middle Eastern investments far more than any single regime's fate.

I think the order question highlight the real bottom line: Beijing doesn’t need the Islamic Republic. It needs a Gulf that stays open for business. If those two things coincide, all the better. If they don’t — and February 28 made the question live in a way it never was before — China’s analysts have to think deeply through which one matters more. The answer will define how Beijing positions itself in whatever Iran emerges from the wreckage.

My take on the issue

It may seem obvious to say, but none of this is comfortable for Chinese policymakers, and the discomfort is visible in the lag between events and response. Beijing does not want to see Iran diminished and Washington in effective control of the Persian Gulf — China’s energy lifelines run through it, and a US-dominated Gulf reshapes the entire strategic geography of the Indo-Pacific competition. At the same time, Chinese officials cannot afford to see their deep investments in the GCC destabilized by the same Iranian partner they refuse to restrain. Previous Gulf frustrations with China, like over the 2019 drone strike into Aramco’s refinery and China’s refusal to condemn Iran, are childsplay compared to the Iranian rocket attacks against the Gulf, including hitting areas near where Chinese companies are operational in Jebel Ali and adjacent zones. The theoretical “what if” scenario is over, and Gulf capitals are absorbing Iranian missile fire. Chinese infrastructure investments, financial exposure, and Vision 2030 partnership deals sit inside the blast radius.

This is Beijing’s paralysis in plain terms. Condemning the US and Israeli strikes too loudly looks like endorsing Iranian aggression against China’s own Gulf partners. Pressuring Iran to stand down looks like capitulating to Washington. So China does neither, and the silence — now softening into carefully worded condemnations of all parties — speaks for itself.

Chinese colleagues will rightly point out that Washington and Tel Aviv are the primary belligerents, and that it is not Beijing’s responsibility to police a conflict it did not start. Any rational analyst must acknowledge this premise, even if they blame China for buying Iranian oil and reportedly supporting Iran’s rebuilding of its ballistic missile program. That argument is not wrong. But it is incomplete. China’s core strategic partners in the Gulf are being targeted by China’s other strategic partner, and Beijing’s chosen instrument of non-interference offers them nothing. The 2023 Saudi-Iran rapprochement that China brokered was a genuine diplomatic achievement — it restored dialogue, normalized communication, reopened trade channels after years of rupture. But it was never a security guarantee. China does not do security guarantees. Its diplomatic model is built on sovereignty, non-interference, and the deliberate rejection of the enforcement role Washington has long played in the region. That model may have rhetorical virtues, but it offers to practical solutions for Gulf security

When Iranian missiles land near GCC infrastructure and Beijing’s response is a statement calling for “all parties to exercise restraint,” China’s Gulf partners notice. The Iran-Saudi deal bought goodwill. It did not buy Beijing the credibility of a power that will show up when it matters. That gap — between China’s economic footprint in the Gulf and its political will to defend the order that footprint depends on — is the defining vulnerability of Beijing’s Middle East strategy, and no amount of infrastructure investment closes it.

This episode will also transform Gulf security perspectives on its relationship with the United States and potential limits that can bring when it comes to reliability (the Arabs sought for diplomacy and Trump brought them war). This episode also showed them that China is also not a reliable alternative. But, for China, this is fine. They never tried to be.

Subscribe now

On February 28, Iranian ballistic missiles struck targets across the Gulf Cooperation Council states hitting the UAE, Bahrain, Qatar, Kuwait, Oman, and Saudi Arabia in retaliation for joint U.S.-Israeli strikes on Iran, which killed Tehran’s Supreme Leader. The Gulf states did not choose this fight. They poured enormous diplomatic capital into preventing it. But they absorbed its consequences, and those consequences landed squarely in the middle of the most ambitious AI infrastructure buildout outside the United States. The ongoing conflict between Israel, the United States, and Iran underscores what the commercial optimism of the past two years in the GCC often wants to forget, namely that Gulf AI ambitions are inseparable from the stability of the broader Middle East, and that stability can collapse overnight.

Subscribe now

The stakes are already operational. The GCC data center market reached $3.48 billion in 2024 and is projected to hit $9.49 billion by 2030. Last May, the United States and UAE unveiled a 5-gigawatt AI data center campus in Abu Dhabi — the largest outside the United States, eventually spanning 10 square miles. Abu Dhabi has committed $3.54 billion in digital infrastructure through its Digital Strategy 2025–2027 and aims to become the world’s first fully AI-native government by 2027. The TAMM platform, powered by Microsoft Azure and G42, already delivers 950 government services to 2.5 million residents, processing over 10 million transactions a year — healthcare access, license renewals, utility payments, traffic fines. These systems serve real populations in real time. This week, they sat directly in the path of Iranian missiles following Israeli-American strikes.

Assessing the Physical Stack

Some in Washington have highlighted that Gulf AI deployment raises important questions that serve as a necessary starting point for assessing the physical security risks facing AI infrastructure in the region. Mona Yacoubian and Samuel Zabin at CSIS argued this week that regional adversaries could target data centers, energy infrastructure supporting compute, and undersea cable chokepoints just as they have long targeted petroleum facilities in the UAE and Saudi Arabia. Sam Winter-Levy, with Chris Chivvis, previously warned in the Washington Post that drone threats make the UAE and Saudi Arabia dangerous homes for critical AI data centers. Winter-Levy has also cautioned repeatedly that concentrating U.S. compute capacity in the Gulf introduces strategic vulnerabilities that policymakers have not adequately addressed.

Their analyses provide a good starting point for assessing the security risks to AI infrastructure, especially the risk of exposure to kinetic strikes. The unfolding conflict in the Persian Gulf moves the conversation from theoretical to an empirical reality now that Iranian missiles and weaponized drones have targeted all of the Gulf countries, with the exception of Oman. Airspace is closed across the Middle East. Maritime chokepoints are now under threat of closure, including the Straits of Hormuz and potentially, the Bab al Mandeb. What does the current conflict reveal about the specific vulnerabilities in the physical AI stack and what must change in how the Gulf builds, defends, and operates its future AI ecosystem?

Lessons from the Current Conflict

Defense integration must come first, not last. The most consequential lesson from this week has been the resounding importance of defense integration across the GCC. The UAE intercepted Iranian ballistic missiles with its air defense systems. Qatar successfully thwarted multiple attacks targeting its territory. Gulf states possess sophisticated, layered air defense architectures, and those systems performed under live fire. The physical AI stack is no more inherently vulnerable to ballistic missiles than an oil refinery or a desalination plant, both of which Gulf states have long defended as critical national infrastructure. The operational question is whether data centers are being integrated into those existing defense umbrellas or treated as commercial real estate outside the security perimeter. A 5-gigawatt campus housing hundreds of thousands of advanced Nvidia chips, processing sovereign government data, and delivering public services to millions of residents is a national security asset which demands the same dedicated layered defense — missile interceptors, counter UAS systems, and more — that Gulf states already deploy around their energy and water infrastructure. Defense integration cannot be viewed as a “nice to have”. For the physical stack to survive the threat environment this week laid bare, it has to be a foundational design principle integrated into the stack.

Concentrated infrastructure invites concentrated risk. The economics of hyperscale AI tend to push states toward establishing AI hubs which concentrate massive power supply, cooling capacity, and fiber connectivity co-located in a single campus to maximize efficiency. The Abu Dhabi mega-campus is a good example. But overconcentration raises the potential that a single strike can create layered damage to the AI stack. A successful strike on a hyperscale data center complex running critical government inference workloads would both destroy critical hardware but potentially sever the digital layer through which states deliver public services, process sensitive commercial transactions, and manage sensitive government data. Abu Dhabi’s TAMM platform has already reduced offline government interactions by 90 percent. If the AI systems behind that platform go down, the government’s ability to serve its population degrades in real time. That same lesson goes for all national AI systems employed in GCC countries. This requires more strategic thinking about the design and construction of AI infrastructure design, including geographically distributed inference nodes, mirrored workloads, and the ability to failover critical services to out-of-theater backup in minutes if there is a targeted strike. Redundancy is important, expensive, and non-negotiable.

Connectivity chokepoints are as vulnerable as the facilities they serve. A data center without network connectivity is an expensive, air-conditioned room full of expensive silicon technology. The Gulf’s fiber optic infrastructure funnels through a narrow set of maritime corridors — the Strait of Hormuz and the Bab al-Mandab — that are now under active threat. Four undersea cables in the Red Sea were cut in 2024, disrupting roughly a quarter of internet traffic between Asia, Europe, and the Middle East. The potential for sabotage of undersea infrastructure are likely to become higher risks the more important AI becomes in the Gulf.

Even if every data center in the Gulf survives a future conflict unscathed, severing or degrading the connectivity infrastructure that link them to global networks and to each other would cripple inference delivery, isolate cloud environments, and interrupt real-time data flows critical for the continuity of AI-powered services. The physical stack covers the entire network topology from facility to end user. Redundant overland fiber paths, satellite-based backup connectivity such as low-earth orbit constellations, and diversified cable landing stations are critical for any AI deployment to be resilient against the kind of escalation the Gulf is experiencing at the moment.

Energy supply must be treated as a wartime planning problem. Data centers consume enormous quantities of electricity, and the Gulf’s pitch to hyperscalers rests on abundant, cheap energy. But energy infrastructure in the region has been a target of Iranian and proxy attacks for years, from the Abqaiq strikes in 2019 to repeated Houthi attacks on Saudi facilities throughout the Yemen war. Backup generators at a data center may buy hours of continuity, but they are a poor substitute for grid access during a sustained military campaign, which is precisely what the Gulf now faces. The 5-gigawatt Abu Dhabi AI campus alone would consume enough electricity to power a major city. That is not a facility which can run on diesel. If the power grid feeding a hyperscale campus suffers power losses, the disruption cascades directly into the AI services running on that campus, then from there into the government functions, commercial operations, and civilian services that depend on uninterrupted inference. Gulf states normally plan energy resilience for their petroleum exports in terms of strategic reserves measured in months. They should apply the same logic to the power systems supporting their AI infrastructure. In other words, they will need to establish dedicated generation capacity, fuel reserves, and physical separation between the energy assets serving data centers and those serving broader industrial grids, so that a strike on one does not cascade into the other.

The security environment is structurally volatile, and AI architecture must internalize that reality. U.S. commercial actors exploring the Gulf as an off-shore option for U.S. data centers must now come to terms that the physical stack cannot be designed around assumptions of sustained regional stability. All of the commercial optimism of 2025 generated by Trump’s Gulf tour, the mega-deals, the sovereign cloud announcements unfolded in a matter of hours. Any AI infrastructure built on the assumption that the security environment has permanently improved will prove as fragile as the diplomacy that preceded this week’s strikes.

Gulf AI planners need strong baseline requirements for operating in a volatile region. Critical AI facilities should be built to survive kinetic strikes, not just cyberattacks. AI workloads may need to be run across multiple sites distributed across a geographic region, so that destroying one facility degrades service rather than kills it. Essential government functions which currently rely on AI systems — visa processing, healthcare, emergency response, financial transactions — should retain traditional non-AI backup systems, so that a prolonged outage does not leave millions of residents unable to access basic services. And Gulf states should embed legal frameworks in their existing contracts with hyperscalers and cloud providers that allow critical AI systems to be shifted to secure facilities in allied countries within minutes of a disruption — locking in the technical infrastructure, data transfer protocols, and operational authorities needed to execute that shift before a crisis, not during one.

Built to Survive

The ongoing Iran conflict and its deadly impact on the GCC does not invalidate Gulf AI ambitions. But it is a stark reminder that regional threats must shape how these states build their AI ecosystems, not just what they build. The Aramco precedent cuts both ways. The 2019 attack demonstrated that concentrated critical infrastructure in the Gulf can be struck with devastating effect — drone and cruise missile strikes knocked out 5.7 million barrels per day of Saudi crude in a single morning. But Aramco’s recovery showed that redundant capacity, operational depth, and strategic reserves can contain and reverse the damage.

The AI sector needs to absorb both halves of that lesson. Gulf states know their threat environment. They have worked with the United States to defend their territory under fire for decades, and their air defense systems proved themselves again this week. The challenge now is to ensure that the digital infrastructure they are building commands the same strategic protection they have long afforded their energy assets. If compute is truly the new oil, it deserves the same defenses.

Subscribe now

When Dubai built Jebel Ali in the 1970s, the UAE did not need a port that large. The domestic economy was not yet ready to absorb the volume of trade it was sized to receive, and the Gulf’s population was a fraction of what it is today. The project was met with widespread skepticism. What the late Sheikh Rashid bin Saeed Al Maktoum envisioned was an entrepôt — a transit hub capturing economic value not by producing goods but by controlling the node through which others must pass. Jebel Ali became one of the world’s busiest ports not because the UAE is large, but because Sheikh Rashid understood that control over strategic trade nodes have a compound effect. The rents, economic ties, and leverage it generates accumulate and reinforce each other across decades creating political, economic, and financial capital. Dubai’s current status reflects the bet paid off. The UAE did not need to manufacture the goods moving through Jebel Ali, it just needed to create the infrastructure needed to be indispensable to those who did.

Subscribe now

The Gulf states are now applying that logic to artificial intelligence. The data centers rising across Abu Dhabi and Riyadh are not built to serve only the 60 million GCC residents. They are being built to position the GCC as the world’s AI entrepôt, the pathway through which three to four billion people across South Asia, Southeast Asia, and Africa access frontier AI inference. While some nations are concerned about getting stuck between the U.S. and China over AI, the GCC may just create the third option.

The Inference Layer as Strategic Terrain

Much of the discourse on GCC AI investment is organized around a potentially misleading frame. Which state is spending more, which national champion has secured the better NVIDIA partnership, whether Abu Dhabi or Riyadh will claim regional AI primacy — these questions matter, but they risk obscuring the more consequential one: what the Gulf is actually building. Gulf states are not trying to out-train OpenAI. Saudi Arabia and the UAE want frontier model capability of their own, and American models underpin most Gulf AI innovation today — a dependency unlikely to disappear soon. But the pursuit of AI sovereignty may ultimately prove less about developing frontier models than about owning the infrastructure through which frontier AI will likely reach much of the Global South. In doing so, the GCC risks positioning itself — whether it intends to or not — as the node of structural intermediacy between the world’s two dominant AI powers.

The distinction between training frontier models and deploying them at inference is important. Training frontier models require concentrations of compute, talent, and capital that only the United States and China can currently sustain at the frontier. The UAE and Saudi Arabia want to make the GCC into the third AI hub. Inference, meanwhile, is the deployment of an existing frontier model across a range of functions and outputs at scale — a state running Anthropic’s Claude or DeepSeek to streamline government services, financial systems, healthcare infrastructure, or logistics networks. This is where economic and political value accrues to end users. The capital requirements, while large, are achievable for middle powers with sufficient financial reserves. The strategic upside, however, may be disproportionate to that investment.

Whoever hosts the inference layer determines which models run, whose data they process, and under what governance terms. For a nation like Qatar or Kuwait, with large capital reserves, controlling that layer may reinforce domestic oversight of AI usage within their own borders. For a nation that lacks inference infrastructure entirely, utilizing any frontier AI model for inference means accepting another state’s influence over all of those factors simultaneously. That is how AI infrastructure creates structural power over how artificial intelligence mediates the decisions of the developing world. The current options for AI inference deployment at scale are largely China and the United States. The U.S. restricts who has access to its chips, infrastructure, and frontier models to a few key partner countries, while China openly offers its models and infrastructure to help countries scale AI in their own countries. Most countries may want U.S. AI stacks, but be unable to access them. Thus, they only have a second-best, but still strong China option. The GCC may be able to position itself as the third option.

The geographic case for Gulf AI intermediacy seems compelling on its own terms. The region sits at the intersection of submarine cable routes connecting Europe, South Asia, Southeast Asia, and Africa. Governing institutions across those same regions — government ministries, central banks, healthcare systems, enterprises managing sensitive national data — face a common dilemma. Routing AI workloads through Chinese or American infrastructure creates a different but symmetrical set of dependencies. Many of these states have enacted or are building data sovereignty frameworks that require sensitive workloads to be processed outside great power jurisdiction — yet they lack the domestic infrastructure to fulfill that requirement independently.

The GCC pitch is that a credible Gulf inference entrepôt could resolve the dilemma. Gulf data centers may offer countries across Africa and the Middle East affordable, high-quality AI infrastructure — likely built on American models and aligned with U.S. standards — including systems with meaningful Arabic-language capability, processed outside the United States and China, with the geographic proximity to deliver real performance advantages to the markets that matter most. This, however, may not be acceptable to U.S. lawmakers if it exposes U.S. systems to potential compute arbitrage by U.S. rivals who want access to U.S. systems without needing physical hardware.

Node Control and Weaponized Interdependence

This position is best understood through the framework of weaponized interdependence, a concept developed by Henry Farrell and Abraham Newman. Their core argument is that global economic networks grow asymmetrically. Certain nodes — the hubs through which disproportionate flows of goods, capital, or data pass, like ports, financial clearing houses, or internet exchanges — accumulate far more connections than others. Whoever controls those nodes holds disproportionate power over everyone who depends on them. That power operates through two mechanisms. The panopticon effect describes the capacity to monitor traffic flowing through a controlled node, enabling the controlling actor to gather intelligence on who is doing what and with whom. The chokepoint effect describes the capacity to restrict or shut down access entirely, cutting off any actor who steps out of line. Both dynamics are structural leverage endemic to AI inference infrastructure. Whoever governs the node through which states access frontier AI models can see what governments are building, what institutions are automating, and what populations are asking. And, it offers a kill switch to sever that access entirely if the political moment demands it.

This is why the U.S.-China competition for Gulf AI infrastructure is critical. A realist view of international relations would assert that AI deployment is ultimately about advancing national interest for each actor. The U.S. deploys its AI stack to the Gulf to secure strategic advantages as the foundation of the GCC’s AI ecosystem — crowding out Chinese alternatives in the process. The GCC, in turn, deploys it to build advanced AI systems which it can leverage to entrench its position as the indispensable node between great power AI ecosystems and the developing world. A genuinely neutral Gulf AI entrepôt sitting between Gulf-deployed American models and other state users would not merely redistribute commercial rents — it would intermediate, and potentially neutralize, the panopticon effect that current U.S. AI dominance confers. A hub that processes the AI workloads of Global South governments and enterprises outside American or Chinese jurisdiction denies both powers the informational advantages that direct infrastructure control provides. The Gulf’s AI build is, from this vantage point is both a technology investment as well as a bid for a new form of geopolitical leverage, one that neither great power is inclined to concede without extracting significant concessions in return.

The picture is more complicated for Washington than for Beijing. Gulf AI infrastructure is set to run predominantly on American models, chips, and cloud architecture — which means the U.S. retains a degree of visibility and leverage that a Chinese-built alternative would not preserve. But that dependency cuts both ways. The Gulf needs American technology to build credibly at scale. Washington needs Gulf infrastructure to extend its AI ecosystem into markets it cannot reach directly. The result is less adversarial than interdependent. What it also means, however, is that the Gulf’s claim to genuine neutrality is qualified from the start — a hub built on American technology is never fully outside American reach, whatever its geographic address suggests. That has ripple effects on how AI sovereignty is defined in the long run.

The Conceptual Third Option

China may ultimately be the winner across many parts of the world where U.S. AI infrastructure and technology remain inaccessible. If Washington does not permit the export of its AI tech stack to Africa, the Middle East, and Central Asia — even at the inference layer — Beijing will enter those markets as the frontrunner. It is arguably already there. For countries concerned about locking into Chinese infrastructure, however, a conceptual Gulf AI model could represent an acceptable third option: regionally grounded, Arabic-capable, and governed outside the jurisdiction of either great power.

A Gulf-trained frontier model — built on deep regional data, fine-tuned for Middle Eastern legal, financial, and cultural contexts — could offer other countries something neither Washington nor Beijing currently provides: high-performance AI that is linguistically and culturally proximate, priced for markets that American and Chinese hyperscalers have historically underserved, and governed on terms that smaller states may find more acceptable.

China’s AI advantage in the developing world rests largely on governance flexibility and price — not necessarily model quality or cultural fit. A credible Gulf frontier model could contest both the markets where Chinese infrastructure is already entrenched and the markets where it has not yet taken hold. The GCC would not be channeling American AI models to the developing world, but instead offering its own built on American hardware.

The dependency question does not disappear entirely. Gulf frontier models would still be trained on American chips, and that upstream relationship preserves U.S. leverage. But the nature of the dependency changes. An Gulf inference hub running on American models is structurally subordinate to U.S, leverage. A Gulf state like Abu Dhabi running its own frontier model on American chips has greater autonomy over what the model does, whose data it trains on, and under what terms it is deployed. The chips remain American. The AI, increasingly, would not be.

Whether Saudi Arabia and the UAE can execute on that ambition at the required scale and timeline is an open question. Developing a frontier model requires not only compute but also talent, data infrastructure, and institutional capacity that both states are actively seeking to build. That direction of travel is pretty clear. If Trump’s approval of larger volumes of advanced chips materializes and Gulf frontier model development follows, the third option moves from conceptual to competitive.

Understanding Gulf Leverage

Saudi Arabia and the UAE’s leverage in the AI race rests on gaps in America’s own infrastructure gaps, namely energy. The U.S. faces a projected AI data center power demand of 130 gigawatts by 2030 that substantially exceeds its available domestic generation capacity under current development timelines. GCC countries are betting that a significant share of U.S. AI computation infrastructure will migrate offshore, and they want to capture it. The Gulf’s structural advantages — cheap and abundant energy, desalination capacity for data center cooling, established connectivity infrastructure — make the region an attractive location for AI infrastructure and functionally necessary for sustaining American AI ambitions.

Much of this relies on GCC interstate coordination. Mohammed Soliman argues that the region should focus on the development and expansion of energy infrastructure with the objective of achieving grid and data integration to meet AI demand — a unified framework that would sharpen the Gulf’s potential to emerge as the third AI hub globally. The geopolitical realities of the GCC, however, often lend toward internal competition, particularly between Riyadh and Abu Dhabi. That rivalry is not merely diplomatic — it shapes investment priorities, infrastructure decisions, and the degree to which Gulf states are willing to subordinate national AI ambitions to a collective regional strategy. Otherwise, both nations will be racing to develop their own AI entrepôt which, if realized by either or both, could create more optionality for Global South customers but also greater fractionation of AI models and infrastructure across the region. A unified GCC AI framework remains the condition under which the third option becomes most credible.

Where the AI Entrepôt Argument Is Contested

I find the entrepôt thesis is analytically compelling, but it rests on assumptions that deserve direct scrutiny. Four vulnerabilities stand out.

First, AI models are not necessarily neutral. By nature, they embed the legal assumptions, cultural frameworks, and epistemic priorities of their training context. The Gulf’s entrepôt position could be reframed as a question of whose non-neutrality to host.

Secondly, there are limits of the Farrell-Newman framework as applied here. Weaponized interdependence was built to explain how great powers exploit nodes — not how middle powers use nodes to resist great power pressure. The historical cases cut predominantly in the other direction. The Gulf is attempting something the framework does not strongly predict is achievable: durable autonomy at a contested hub between two competing hegemons. The G42 divestiture from China is the first data point, and it does not favor the neutrality thesis.

The third concerns Global South demand. The piece’s argument presupposes that Global South governments and institutions are actively seeking an AI inference host outside great power jurisdiction. That demand seems to be plausible, but maybe not to the scale necessary that financing a Gulf “third option” might require. Many countries in recent years have gone directly to Chinese providers because either the U.S. option was not available or states found the Chinese option for accessible due to price and fewer governance constraints. The intra-GCC dynamic introduces a further complication. Both Saudi Arabia and UAE are moving firmly into the American orbit. If both states build major inference infrastructure with divergent great power alignments, the singular Gulf entrepôt thesis fractures into two competing entrepôt nodes — undermining the collective leverage I outlined above

Finally, there is the question of where AI itself is heading. The entire entrepôt model rests on an implicit assumption: that AI inference will continue to require large, centralized data centers that smaller states must access through regional hubs. That assumption may not hold. The broader trajectory of AI development points in the opposite direction — toward smaller models, open-weight architectures, and deployment on modest local hardware. If inference increasingly migrates to the device or institutional level, the demand for centralized Gulf infrastructure could erode before it reaches full utilization. The Gulf’s entrepôt bet is, in part, a bet on continued centralization. That is a reasonable bet today. It is less obviously reasonable across the decade-long investment horizon the current buildout requires.

Conclusion

The Gulf does not need to train the world’s most capable models to occupy a position of structural consequence in international relations. The entrepôt logic requires only that the world cannot route around it. But the weaponized interdependence framework identifies a structural problem for Gulf Arab states as middle powers. Building an AI hub between two competing great powers does not mean the GCC transcends the existing competition — on the contrary, it may become the competition’s object. Both Washington and Beijing face structural incentives to ensure that Gulf AI infrastructure serves their respective panopticon interests: running their models, adopting their governance frameworks, excluding the other’s hardware and firms. A Gulf AI entrepôt that maintains genuine neutrality may be seen as strategically threatening by both great powers simultaneously. The G42 divestiture represents the first visible instance of this pressure succeeding at scale. How long Gulf neutrality can last, especially as Washington and Beijing compete for global adoption of their respective AI ecosystems under Trump’s new AI export push, remains the central unanswered question hanging over the entire enterprise.

Subscribe now

The more I speak with aid providers, Arab officials, and potential actors in reconstruction, the more complex Kushner’s ambitious plans get. The biggest impediment is arguably simplest.

Subscribe now

The Trump administration’s vision for Gaza is ambitious. At the World Economic Forum in Davos in January, Jared Kushner unveiled a “master plan” for “New Gaza”—a phased reconstruction featuring more than 150 skyscrapers, a seaport, an airport, data centers, and coastal tourism zones, all projected to cost upward of $25 billion. The Board of Peace, signed with fanfare by some two dozen countries, aims to create a new chapter in which Arab states, international investors, and American leadership converge to rebuild the strip and reshape the region. But no matter how bold (and divisive) the blueprint, Trump and Kushner face the same obstacle that has frustrated humanitarians, the United Nations, and the Biden administration since October 7, 2023: Israel’s ironclad control over what enters Gaza.

Throughout the war, Israeli authorities blocked the entry of goods so basic they do not remotely register as dual-use concerns. A CNN investigation in March 2024 reviewed documents compiled by major participants in the humanitarian operation and identified the items most frequently rejected by Israeli inspectors, which included anesthetics and anesthesia machines, oxygen cylinders, ventilators, water filtration systems, dates, sleeping bags, and cancer medications. The Washington Post, after contacting 25 aid groups, UN agencies, and donor countries, reported that blocked items also included crutches, generators for hospitals, maternity kits, solar panels, and—in one memorable case—chocolate croissants, which Israeli inspectors apparently deemed luxury foods inappropriate for a war zone. If a single item on a truck was rejected during inspection, the entire truck was sent back.

The restrictions were never codified in a single published document. For years, the Israeli human rights organization Gisha fought a legal battle under Israel’s Freedom of Information Act to compel the government to reveal its policies on restricted goods. When the list was finally published, it contained 56 items requiring special approval across the occupied Palestinian territory and 61 additional items specific to Gaza—on top of the international Wassenaar Arrangement’s standard dual-use categories. But as Gisha documented, the categories were so broad—“vehicles,” “communications equipment,” “communication-support equipment”—that thousands of individual items effectively fell within their scope. When Human Rights Watch filed its own freedom of information request during the war, Israeli authorities said they were still using a dual-use list published in 2008. Gisha’s executive director, Tania Hary, told Human Rights Watch that authorities were “interpreting the list very broadly, which is nothing new, except it’s taking place on the backdrop of a humanitarian catastrophe.”

In practice, aid workers described something more akin to a shadow regime. One senior humanitarian official told CNN the system was “deliberately opaque, deliberately ambiguous.” Another described fighting “a different war” at Egypt’s Rafah crossing—a war to bring humanitarian aid into Gaza. Human Rights Watch researchers who visited Egypt’s North Sinai region in April 2024 spoke with workers from 11 UN agencies and aid organizations; all said Israeli authorities continued to obstruct entry, rejecting entire truckloads in an ad hoc manner with no explanation or possibility of appeal. The Biden administration pushed against these restrictions, failed, and ultimately acquiesced. What began as wartime exceptionalism became the status quo. By the time the ceasefire took effect in October 2025, the infrastructure of restriction—the inspections regime, the informal veto list, the bureaucratic inertia—had not been dismantled. It had simply been inherited by a new administration with far grander plans.

And those plans demand vastly more than tents and water filters. Kushner’s reconstruction vision requires concrete, steel, cranes, excavators, telecommunications equipment, and eventually semiconductors and advanced electronics—categories that sit squarely within the restricted zones of Israel’s dual-use framework. Before a single smart city can rise, Gaza’s estimated 68 million tons of rubble must be cleared. That alone demands heavy machinery that Israeli authorities have never permitted to enter the strip. Prefabricated housing structures—the minimum necessary to shelter displaced Palestinians while longer-term construction proceeds—require materials and components that fall within the categories Israel has historically blocked. The gap between what the Board of Peace promises and what Israeli border controls currently allow is not a crack. It is a chasm.

A January 2026 investigation by The Guardian revealed that Israeli authorities were already operating a parallel import system—allowing commercial traders to bring in goods classified as dual-use, including generators and metal pallets, through the same checkpoints that block such items for humanitarian organizations. The items were being sold openly in Gaza’s markets. As one diplomatic source told The Guardian, “It seems highly improbable that the Israelis don’t know about them.” If Israeli authorities can permit commercial imports of restricted goods while blocking humanitarian deliveries of the same items, the political dimensions of these restrictions become impossible to ignore. The system is not merely a security apparatus. It is a lever of control.

Israeli officials are unlikely to reverse course voluntarily. The security establishment views restrictions on materials entering Gaza as a core element of its containment strategy, and the political incentives within Israel’s governing coalition reinforce a maximalist posture. Finance Minister Bezalel Smotrich has called for Israel to establish a military government in Gaza and take unilateral responsibility for the strip’s future—a vision fundamentally incompatible with Kushner’s reconstruction timeline. Even a marginal relaxation of import controls would require sustained diplomatic pressure from Washington—pressure the Trump administration has so far directed at selling the plan rather than confronting the operational realities of implementing it.

The devil, as always, is in the details. Grand Gaza visions must eventually become purchase orders, shipping manifests, and trucks queuing at Rafah and other borders. The next phases of the ceasefire deal—including Hamas disarmament—are likely to drag on for months, if not longer. Kushner himself acknowledged at Davos that reconstruction would proceed only in portions of Gaza where Hamas had demilitarized. But Hamas officials have said they would consider “freezing” their weapons only in the context of achieving Palestinian statehood—an Israeli nonstarter. Until meaningful disarmament occurs, Israeli authorities will almost certainly maintain draconian controls over what crosses into Gaza, arguing that construction materials could be diverted for military purposes. This creates a paralyzing circular reasoning, Reconstruction cannot proceed without relaxed border controls, but Israel will not relax border controls without disarmament, and disarmament depends in part on demonstrating that the international community will deliver on its reconstruction commitments.

This bottleneck threatens to unravel the broader regional architecture the Trump administration has constructed around the Board of Peace. Arab states—Saudi Arabia, the UAE, Qatar, Egypt—signed onto the initiative with the expectation that it would produce tangible results for Palestinians. If Israeli border restrictions prevent reconstruction from materializing, those governments will face domestic pressure to distance themselves from a process that delivered only a nice ceremony but no concrete change. Hamas, meanwhile, will seize on any failure to deliver reconstruction as evidence that the international community reneged on the terms of the ceasefire. And Israeli authorities are reportedly considering unilateral military action if they concludes that the diplomatic process has failed to neutralize the security threat from Gaza—a decision that would shatter the ceasefire entirely.

Signs of friction are already visible. Israeli officials expressed persistent skepticism about the Board of Peace before and during its launch. Netanyahu’s office initially criticized the makeup of the board’s Gaza executive committee, which includes Turkey and Qatar—both of which have backed Hamas. Israel did not send a representative to the signing ceremony in Davos. Most of Washington’s European allies have declined to join. At the ceremony itself, Kushner felt compelled to address the skeptics directly: “Just calm down for 30 days,” he urged. “The war is over. Let’s work together.” That a public plea for patience was necessary at the plan’s unveiling tells you something about the distance between Washington and Jerusalem. As of early February, Israel had still not formally signed the Board of Peace charter, despite Netanyahu’s announcement that it would join.

Trump does not tolerate being made to look like a fool. If Israeli restrictions on border access become the visible reason his Gaza plans stall, the political dynamics could shift in unpredictable ways. A president who has positioned himself as Israel’s greatest ally may find himself in an uncomfortable confrontation with a government that refuses to give his plan the operational space it needs. The reverberations would extend far beyond the bilateral relationship. Arab states that took a political risk by endorsing the Board of Peace would reassess their exposure. Regional actors would recalibrate their expectations of American leverage. And the window for a negotiated resolution in Gaza—already narrow—would close further.

This is the lesson of the past two years: the most consequential decisions about Gaza’s future hinge on what Israeli authorities allow to enter Gaza. The Biden administration learned this the hard way. The Trump administration, for all its confidence and ambition, is now running headlong into the same reality. Until Washington is willing to expend serious political capital to change Israeli behavior at the crossings, no plan for Gaza—however visionary—will survive contact with the ground.

Subscribe now

Artificial intelligence competition in the Middle East is often framed as an infrastructure story. Data centers are being built in the desert. Chip access is being negotiated at the highest levels. Sovereign wealth funds are sending billions toward Silicon Valley to build their own AI infrastructure based on U.S. models, and using hyperscalers to advance their AI ecosystems.

Most attention from observers is often on the physical hardware of AI ecosystems in the Middle East, largely because this is where the most consequential decisions are made regarding long-term access to technology. If the UAE and Saudi Arabia are successful in transforming their countries into AI hubs, this will be an important step for the Middle East. But, a lesser known competition I find interesting is the question of indigenous “Arab AI”. As the Middle East nations move to catch up to the rapidly advancing global AI race, how will 400 million Arabic speakers interact with artificial intelligence, and how does AI itself come to understand the Arab world?

Arabic remains one of the most significant unsolved problems in natural language processing. The language spans 30+ dialects across 22 countries, features morphological complexity that defeats standard tokenization approaches, and carries cultural and religious context that generic models routinely mishandle. According to recent analysis, Arabic is spoken by over 491 million people across 22+ countries—yet only 0.5% of natural language processing (NLP) research focuses on it. This gap creates an opening that multiple actors have recognized, though they’re approaching it in fundamentally different ways, with different implications for the region.

Hardware over software?

The dominant AI narrative in the Middle East focuses on hardware. Last year, the Commerce Department approved 70,000 advanced chips to the UAE and Saudi Arabia, conditioned on Gulf partners distancing themselves from Chinese technology firms. Meanwhile, American hyperscalers like Amazon, Microsoft Azure, Google Cloud and Oracle Cloud have been welcomed by the region and are building data centers across the region and partnering with the region’s top AI companies, including the UAE’s G42 and Saudi’s Humain. China still maintains a foothold, but it currently is dwarfed by the GCC’s increasing lock-in to American innovation.

The hardware story matters because compute is a prerequisite for the Arab Gulf’s serious AI development. But the hardware infrastructure does not solve the Arabic problem. The flagship models from American companies remain English-first systems with Arabic capability added on. They can translate and respond in Arabic, but they reason through conceptual reasoning and thinking architecture developed in English and trained predominantly on English-language data. When these models process Arabic queries, they’re performing sophisticated translation rather than native comprehension.

Compute alone doesn’t produce AI systems that understand the Arab world on its own terms. A Saudi researcher querying an American model about regional dynamics receives answers filtered through frameworks that reflect how Silicon Valley conceptualizes the world—not how Riyadh does. The same goes for a UAE official querying a Chinese model. If the roots of Arab AI are non-Arab (assuming they are built and trained Chinese or American models), when does an AI model become distinctively “Arab” in thought, reasoning, and language?

The Gulf’s Arab LLM vision

The Gulf states are taking on the challenge. Both Saudi Arabia and the UAE have launched serious indigenous Arabic AI initiatives.

Saudi Arabia’s SDAIA developed ALLaM, training it on a 500-billion-token Arabic dataset—the world’s largest—assembled by mobilizing 16 government entities. Over 400 subject matter experts tested the model through more than a million prompts. The resulting system, now deployed through HUMAIN Chat, explicitly encodes Islamic values and regional cultural context. Saudi officials describe it as “sovereign AI”—built in the Kingdom, by Saudi talent, for Arabic speakers.

The UAE’s Jais, developed by G42’s Inception unit in collaboration with the Mohamed bin Zayed University of Artificial Intelligence, takes a different approach. The latest Jais 2 release, built with 70 billion parameters and trained on the largest Arabic-first dataset, covers a large range of Arab dialects and aims to accurately reflect and respond appropriately to the cultural norms, values, and references of Arabic-speaking communities. The developers have gathered conversational datasets across regional dialects, enabling the model to respond in Lebanese Arabic to Lebanese users, Gulf Arabic to Gulf users, something other global models cannot yet do. Why is that important? If you ask ChatGPT to translate a text passage into Syrian dialect, it can do a surprisingly good job. But, it is required to translate from English to Syrian Arabic through a complex algorithm. Imagine the difference if Arab AI has a strong Arabic ontological and context layer which can think and reason in Arabic?

There are structural challenges. Researchers from the Association of Computing Machinery (ACM) note that “...limited regional collaboration and infrastructure across Arabic-speaking countries continue to hinder large-scale development. While resource-rich nations have invested in AI research, the absence of a cohesive research network and weak industry-academia integration prevent widespread progress.”

As they note, resource-rich nations are investing in AI, but are they investing in Arab AI? This may be one of important turning points in enhancing AI diffusion for Arabic-speaking users in the Arab world.

China’s growing role

While the U.S. has been particularly restrictive (at least until the Trump administration) on the hardware side, Chinese AI firms have taken a markedly different approach. Instead of just focusing on hardware (where they are losing to the United States), Chinese firms are engaging directly with Arabic linguistic and cultural challenges through sustained research collaboration with key Arab AI institutions.

The most striking example is AceGPT, a large language model built through partnership between Saudi Arabia’s King Abdullah University of Science and Technology (KAUST) and two Chinese institutions: the Chinese University of Hong Kong, Shenzhen, and the Shenzhen Research Institute of Big Data. The project was launched by a Chinese-American professor at KAUST and built on Meta’s open-source Llama 2 architecture. Unlike models adapted for Arabic after the fact, AceGPT was designed from inception with Arabic linguistic and cultural alignment as primary objectives—what the developers describe as training for “cultural sensitivity and alignment with local values.”

The results have been significant. According to the Shenzhen Research Institute, by the end of 2024, AceGPT offered models across multiple sizes (7B, 13B, 32B, and 70B parameters), and “significantly outperforms competitors—specifically the Jais model developed in the UAE—making it the world’s leading open-source Arabic large language model.” The trilingual design encompassing Arabic, Chinese, and English is reflective of the research team’s composition, but also a strategic calculation that Arabic AI development might advance faster through Chinese collaboration than through adaptation of English-first American models.

Huawei has pursued a parallel track through infrastructure. In May 2024, Huawei Cloud revealed a 100-billion-parameter Arabic large language model based on PanGu, originally trained on Chinese language data and then adapted for Arabic. The model launched through Huawei’s new Cairo data center positions Chinese AI infrastructure at the entry point for Arabic speakers across North Africa and the broader Middle East, especially in regions where American tech presence remains thinner and Chinese infrastructure investments more substantial.

The financial ties run deeper still. Saudi Aramco’s venture arm, Prosperity7 Ventures, invested $400 million in Zhipu AI, one of China’s leading generative AI companies and a direct open-source competitor to OpenAI. Zhipu AI’s recently-released GLM-4.7 model is said to both outcompete ChatGPT and rival Claude’s Sonnet 4.5. The investment valued Zhipu at approximately $3 billion and made Prosperity7 the sole foreign investor in China’s flagship effort to build a domestic open-source AI champion.

China’s engagement moves beyond just infrastructure investment and anchors direct partnerships on the linguistic and cultural dimensions that make Arabic AI genuinely difficult. Perhaps this is a move to cement GCC access in a world where Chinese technology still cannot outcompete U.S. hardware, but it offers the Arab world something that the Arab world desperately wants as AI advances: an Arab AI.

But the collaboration is not benign. What appears on the surface as economic cooperation may involve exposing critical national data to Beijing. One observer wrote regarding Saudi Arabia that these deals involve “sharing Saudi Arabia’s national data with Chinese algorithms. In exchange for transferring these technologies, Beijing gains access to data flowing from Saudi smart cities to its oil industries—data that constitutes 21st century raw gold for China.” This is the type of exposure which tends to provoke pushback by U.S. policymakers who oppose giving the most advanced hardware to Gulf Arab partners.

An Arab Ontology?

Both the great power approaches and the indigenous Gulf efforts share a common gap: they focus on language fluency. What makes Arab AI will likely need to factor an indigenous knowledge architecture. Speaking fluent Arabic (and all of its dialects) is necessary, but insufficient for an AI model to genuinely “understand the Arab world”. One challenge is that “understanding the Arab world” is a data question. The model has to be trained on Arabic data, which brings with it all the complexities of the region, including competing theories of governance, religion, history, society, but also poetry, history, world views, and more. That is hard to capture when training an AI model. Meanwhile, “understanding the Arab world” also requires structured ontologies that define which entities, data, and sources matter, how these relate to each other, and how regional dynamics actually unfold.

What does ontology mean in the case of Arab AI? Let’s use a couple of case studies:

If you ask: Tell me about PIF

A language model trained on Arabic text knows that “PIF” refers to Saudi Arabia’s Public Investment Fund. But an ontology would encode that PIF is distinct from the Saudi state, that it operates through dozens of subsidiary vehicles, that its investment decisions reflect both commercial logic and strategic priorities set by the Crown Prince, and that its portfolio companies maintain relationships with specific ministries, regional governments, and international partners. The ontology captures not just what PIF is, but how it functions within a web of entities—sovereign wealth funds, ruling family offices, state-owned enterprises, private conglomerates, tribal networks—that collectively constitute Gulf political economy.

If you ask: Tell me about Fairuz

A language model trained on Arabic text is likely to know that Fairuz is a legendary Lebanese singer. But an ontology would enable the AI model to understand that Fairuz emerged from the Rahbani brothers’ musical compositions, which blended Lebanese folk traditions with Western orchestration to create something deliberately pan-Arab in its aspirations. A strong Arab ontology would connect this historical foundation to the region-wide love and devotion her songs inspired. This includes works that became anthems of Lebanese national identity, while becoming background music for nostalgic instagram posts shared thousands of times in English and Arabic. It would trace how her influence threads through contemporary artists like Mashrou’ Leila’s indie rock in Beirut, Cairokee’s post-revolution anthems in Egypt, Omar Souleyman’s Syrian dabke-electronic fusion that found audiences in Berlin, the UK, and Brooklyn. The ontology captures not just who Fairuz is, but how she functions within a web of cultural production—composers, lyricists, national broadcasting elites who decided what millions would hear (think of Egyptian radio’s dominance through the 1970s), and the generational memory that made her voice synonymous with morning rituals across the Arab world.

Data and sources require their own ontological mapping. AI models need to know which information streams are authoritative for which questions.

• Official Saudi press releases signal government positions but obscure internal debates.

• Emirati business publications capture commercial activity but underreport political dynamics.

• Arabic-language social media reflects popular sentiment but amplifies certain voices over others.

• Lebanese financial journalism offers regional perspective but carries its own biases.

A structured ontology maps these sources to their domains of reliability, their known blind spots, and their interlinking relationships to power centers, which enables the AI model to weight and triangulate information rather than treating all inputs equivalently.

If you ask: What is an Arab

This is the most fundamental challenge: how does an AI model define what “Arab” actually is and what narrative of “Arab” is the correct one to portray? The Arab world is not a monolith, and any ontology must grapple with competing frameworks for understanding what binds—or divides—the region.

It would reason with fundamental identity questions:

• Is “Arab” primarily a linguistic category, encompassing everyone from Moroccan Berbers who speak Darija to Lebanese Christians who code-switch between Arabic, French, and English?

• Is Arab a political identity forged through twentieth-century nationalism, with its centers in Cairo, Damascus, and Baghdad?

• Is it a cultural sphere defined by shared literary traditions, musical forms, and social customs—or a religious civilization where Islam provides the organizing grammar?

While an Arab AI model trained on Arabic text absorbs all these competing frameworks without distinguishing between them. An ontology forces a choice—or at minimum, makes the competing frameworks explicit so that users understand which version of “Arab” the system is reasoning through.

Building such ontologies for the Arab world remains will be a large-scale undertaking. Even if Models can be trained to speak Arabic fluently, without a strong ontology, they are no better at understanding the Arab world than me on my first trip to Jordan as a 17 year old teenager with no Arabic going to work on an archaeological dig. Thus, it requires both a strong Arabic language fluency to achieve a broader diffusion of AI integration in the Arab world as well as the right way of thinking and reasoning in Arabic about the Arab world.

What comes next?

There is a clear interest and demand from the Arab world for an “Arab AI”. The U.S. is overly focused on hardware deployment and cooperation through infrastructure, but seems less interested in the region’s Arabic LLM pursuits. Meanwhile, China has used the Arabic language processing needs as an entry point for deepening AI partnerships. This, however, does not necessarily provide the Gulf Arab states the type of technology they need to build to sustain compute capacity necessary to scale their AI ecosystems (Arabic LLMs are not the only objective). One commentator writing with The National argues that Arabic AI represents “a huge gap and a major opportunity: whoever builds the best models for Arabic will gain a strategic data advantage in a massive underserved market.” Both great powers recognize this. So do the Gulf states themselves.

Several dynamics are important to watch:

First, the AI entity that achieves superior Arabic language capability will gain significant advantages in regional AI adoption, whether through indigenous development, American adaptation, or Chinese collaboration. This capability seems to be genuinely contested, and no actor has established clear dominance.

Second, language capability and knowledge architecture are distinct challenges requiring different investments. The Gulf states have focused primarily on the former through infrastructure investments and hardware acquisition. But they have also advanced the Arab AI model through several regional initiatives, but there is still much work to be done. AI hardware and chips will create power compute capacity, but the development of a stronger Arab ontology and knowledge architecture will be critical for building an AI model for the Arab world.

Third, the Gulf’s indigenous models represent something genuinely new in the space. I AI systems designed explicitly to encode regional values and cultural context. This is not too far off from what Chinese authorities seek to create, but within the context of reinforcing CCP governance and Chinese values. This raises questions about how designers pick and choose what values to keep or cast out. This is the ethical layer which requires concurrent discussion as Arab AI is being developed. Whether HUMAIN Chat’s “Islamic values” alignment or Jais’s dialect coverage translate into competitive advantage remains to be seen, but the attempt itself signals the region has deeper ambitions to build an AI model for broader diffusion across the region. As the Arab world may be majority Muslim, it is not fully muslim and indigenous models will need to account for that diversity.

Fourth, the Gulf Arab states still prefer American platforms despite their Arabic limitations. OpenAI’s ChatGPT holds roughly 90% market share in Saudi Arabia and the UAE. Naturally, users are defaulting to the most capable general-purpose system even when it handles their language imperfectly. This creates a window for Arabic-native alternatives but also demonstrates how quickly that window could close.

In the end, I think discourse on indigenous Arab AI is an important because when (not if) Arab AI is developed, its owners and operators have the potential to shape how nearly half a billion people experience artificial intelligence—and how AI models come to represent a deeply complex, historic, yet beautiful region.

Kai Fu Lee, author of AI Superpowers and a leading thinker on AI in China, summarized the position well in his oped with Arab News:

“It may take time for countries to figure out their strategy for building a sovereign AI. But it is critical for the Arab world to quickly catalyze the creation of culturally appropriate LLMs and build a rich ecosystem to allow AI-powered Arabic apps to blossom.”

Subscribe now

Syria’s experience show that U.S. options to the ongoing Iranian revolution do not end well for the Iranian people. However, Syria’s experience also shows the ongoing Iranian revolution is on a trajectory that, long term, should make Tehran nervous.

Subscribe now

The hope Syrians felt after overthrowing Assad naturally gives way to the deep complexities of post-Assad society. For my Syrian friends, removing Bashar al-Assad fulfilled what millions hoped to see after generations of Assad rule and brutal persecution of large segments of the population. The war exposed the regime’s worst evils, but only after he fled did we glimpse how deep the depravity went. Yes, one individual and one regime can inflict mass suffering on civilian populations. I was among those who wanted to see Assad go after 14 years of working on the Syrian aid response. It would be hypocritical for me to call for Assad’s removal while diminishing the ambitions of Iranians in the streets demanding their regime’s overthrow. Structural differences and similarities between Syria and Iran matter when drawing comparisons.

The Iran revolution looks more like Syria in 2011 than Syria in 2024. Assad had burned his relations with Iran and Russia, and his international backers saw little incentive to defend him after 14 years of brutal civil war. He controlled only a fraction of the country; the rest remained frozen along battle lines with limited escalation due to exhaustion and economic duress. Hay’at Tahrir al-Sham (HTS) and its allied militias broke this status quo by recapturing Damascus and western Syria within two weeks.

The Iranian regime, while weakened after the 12-day war, still maintains military strength, domestic security forces, weapons stockpiles, and the ability to suppress mass demonstrations. Overthrowing the regime would likely require military force, or at least some form of military support for protest movements. While Trump fomented protests and threatened military action if the Iranian regime killed protesters en masse, we have not yet seen any use of force.

When Iranian authorities implemented an internet blackout and called for Iraqi militias to cross into Iran to suppress protests, Iranian officials signaled a degree of arrogance—they calculated they could suppress the protests and that Trump would likely not use immediate force to support protesters. Syria’s government made a similar calculation in 2011-2013 when it used brutal force to crack down on protesters. The threat of external intervention was secondary to the internal threat of regime change.

The West has so far taken a similar approach to 2011 Syria: strong words, some threats, little action. Iran lacks a mobilized and heavily armed opposition capable of overthrowing the regime during a moment of weakness, which means external force would be necessary. But for Trump, as for Obama in 2011-2013, this means committing the U.S. to another Middle East conflict. Trump seems more likely to limit himself to surgical strikes that make a point without changing the status quo.

The collapse of the Iranian regime and subsequent vacuum would likely prompt Israeli forces—perhaps jointly with U.S. forces—to carry out widespread airstrikes across Iran to destroy known weapons stockpiles, bury nuclear facilities, and ensure that remaining arsenals cannot fall into the hands of actors who could threaten the U.S. and Israel. When Assad fell in December 2024, the Israeli air force carried out over 400 airstrikes across Syria, crippling the country’s military infrastructure, destroying weapons stockpiles, and eliminating chemical weapons sites. One factor among several was Israeli authorities’ desire to prevent heavy weapons and chemical agents from reaching militants who could potentially use them against Israel. That wasn’t the only driving logic—Israeli forces also invaded and continue to occupy parts of southern Syria.

Israel’s actions immediately cast it as an aggressor in the post-Assad landscape, and many Syrians saw the strikes as a means to cripple Syria’s military and undermine the new government’s ability to mobilize. If the Iranian regime fell, Israeli forces—perhaps with American support—would likely follow a similar approach, but the scale would be far larger due to the breadth of Iran’s defense infrastructure, industrial base, stockpiles, warehousing, and varied defense assets, including nuclear enrichment sites. This would require Israel, with U.S. support, to sustain aerial operations across the entire country to destroy as many sites as possible and prevent weapons from “disappearing” into the hands of unknown actors, especially nuclear material. Such large-scale strikes across Iran would likely cast the U.S. in a negative light, much as Israel’s post-Assad actions did in Syria. Trump would look less like Iran’s liberator and more like an aggressor—a shift that could turn protester sentiment against the United States.

Trump’s team seems more bent toward using escalation to create diplomatic leverage with the current regime than actually removing it if there is a chance at a deal. Trump faces immense pressure from two competing camps in his administration. First, the Iran hawks want regime change. This group has been active for years and sees no alternative but removing the regime, even if that includes military action. Second, the restrainers advocate using military pressure and escalation to bring Iran back to the table for a deal that resolves the nuclear issue without a major conflict forcing the U.S. back into a regional “forever war.” As highlighted above, overthrowing the Iranian regime by force would plunge Iran and the broader region into chaos that would likely force CENTCOM to redeploy forces to the region. This would be deeply unpopular with Trump’s voter base, which would likely split between those who want to see the Iranian regime gone and those who do not want the U.S. dragged back into another Middle East conflict. This is what makes Iran different from Venezuela. Trump’s actions in Caracas, according to voter polling, seemed to strike a chord among his base, who buy the narrative that arresting Venezuela’s president helps address immigration flows and drug trafficking. No such narrative exists for Iran because the legacy of U.S. military intervention in the Middle East casts a long shadow over whatever gains the Israelis or Iran hawks in the Trump administration would spin as justification for intervention.

The Iranian people likely lose either way. Assad clung to power in 2011 at a steep cost—human life, civilian suffering, and international legitimacy. It took 14 years of war to weaken his regime enough for the military opposition to succeed. If Iran’s current revolution succeeds in overthrowing the regime, Israel and the U.S. would likely strike weapons stockpiles across the country, poisoning whatever popular support for America exists among Iranians. If the revolution fails and the regime stays in power, it faces no real deterrent against using force on protesters. Like Assad, the Iranian regime could walk away with impunity after massacring thousands of its own people—undermining the credibility of U.S. threats and demands that Tehran refrain from harming protesters. The regime already crossed that threshold. Yet Washington has taken no action, though a U.S. carrier strike group now patrols the region and Trump reportedly weighs strikes against Iranian officials responsible for harming protesters. Such strikes might restore credibility to Trump’s threats, but they would not change the status quo.

Trump may yet get a deal, but it will not meet the protesters’ demands. That would mean accepting the current regime and abandoning the Iranian people’s ambitions. But like Syria, this marks the beginning of a major shift in Iran’s future. Even if this revolution fails, its seed has taken root and grown deeper. Syrians waited 14 years to see their revolution succeed, and when it did, it surprised everyone. The Iranian regime watched its Syrian foothold collapse with Assad’s fall—a lesson that will likely deepen its insecurity in the coming months. Killing protesters tends to fuel revolutionary fervor rather than extinguish it, especially in a country where domestic anger has reached unprecedented heights.

Subscribe now

This is the next paper in a series of essays on U.S.-China-GCC technology competition. In this piece, I look at how Trump’s move away of the Sullivan Tech Doctrine and transactional turn on semiconductor exports has shifted U.S. technology strategy toward China. While Washington is focused on preserving chip superiority, it maybe underestimating the leverage Beijing holds at the base of the supply chain: critical minerals and rare earth processing. The result is a strategic imbalance in which the United States risks winning the chip race while losing the material foundations that underpins its frontier innovation.

Subscribe now

In September 2022, National Security Advisor Jake Sullivan announced a fundamental shift in U.S. semiconductor policy. Washington would abandon the “sliding scale” approach that aimed to stay “only a couple of generations ahead” of China in advanced technologies. Instead, the United States would seek to “maintain as large of a lead as possible” in chips that power artificial intelligence and defense systems. This became known as the Sullivan Tech Doctrine: export controls engineered not merely to constrain Chinese progress, but to degrade Beijing’s capabilities over time while widening America’s technological advantage.

Two years later, that doctrine lies in ruins and many are not sure what has replaced it. President Trump in December 2025 announced that Nvidia can sell H200 chips to China in exchange for 25 percent of revenues. This was a major policy reversal. It signaled a fundamental reorientation from Sullivan’s strategic denial policy to transactional commerce. Trump would now be treating advanced semiconductors as negotiable commodities rather than the crown jewels of U.S. national security. The Trump administration rescinded Biden’s AI Diffusion Rule before it took effect in May 2025, calling it “overly complex” and “bureaucratic.” Where Biden erected a three-tiered framework restricting chip exports globally—sorting countries by alliance status and capping access accordingly—to prevent Chinese access through intermediaries, Trump instead is prioritizing bilateral deals to those offering the best terms and the biggest checks for the U.S. economy.

The whiplash is striking. In July 2025, the administration approved sales of Nvidia’s H20 chips to China in exchange for 15 percent of revenues. By August, Trump demanded a 10 percent equity stake in Intel as a condition for CHIPS Act funding. In December, Trump made the H200 model—chips roughly six times more powerful than the H20—available to “approved customers” in China for a 25 percent surcharge. Trump justified this move as beneficial for the U.S. economy and leveraged export controls to create new revenue streams for the Treasury.

Then just weeks later, in January 2026, Trump blocked a $2.9 million Chinese acquisition of semiconductor assets from Emcore, citing unspecified national security concerns. The juxtaposition was difficult to square. He approved billions in H200 sales, then prohibited millions in asset purchases? The likely difference is that the former deal generated revenue and maintains market share, while the other involves direct ownership of manufacturing capability. The distinction is critical, but the inconsistency highlights deeper confusion about what semiconductor policy actually aims to achieve.

One leg of the policy debate has crystallized around competing assessments of the U.S.-China technology timeline gap. David Sacks, Trump’s AI and crypto czar, warned in June 2025 that China is “at most two years behind” the United States in semiconductor design, with Huawei “moving fast to catch up” and Chinese firms “adept at evading US export controls.” Sacks pointed to DeepSeek’s breakthrough AI model as evidence that Chinese capabilities are “only months behind” rather than years, making export restrictions futile and counterproductive. Nvidia CEO Jensen Huang called export controls “a failure,” warning that restrictions “gave [Chinese companies] the spirit, the energy and the government support to accelerate their development.” Huang noted that Chinese competitors “are quadrupling capabilities every year” while Nvidia lost access to what he described as “the $50 billion China market,” as the company’s market share plummeted from 95 percent to zero.

Defenders of export controls argue the policies are working. While China’s SMIC has manufactured 7nm chips through costly multi-patterning techniques, these chips are produced at low yields and high costs, leaving China still reliant on Dutch, Japanese, and U.S. equipment it cannot yet replicate domestically. New reporting suggests that China is gradually making breakthroughs, especially in its lithography machinery, to bridge this gap. Intel CEO Pat Gelsinger argued at Davos 2024 that China faced a “10 year gap and a sustainable 10 year gap” as a result of U.S. export control policies. The gap is not closing, Sullivan’s camp argues—it is widening, and loosening controls now would squander the time gained by U.S. restrictions.

Trump’s New Approach to AI Superiority

Both Sullivan and Sacks share a core assumption: maintaining American semiconductor advantage is what matters most. Sullivan pursued this through strategic denial to degrade Chinese capabilities over time. The Trump Administration has taken a similar approach to the Biden administration’s “tall fences, little yards” approach under State Department’s new Pax Silica alignment—a strategic partnership network announced in December 2025 bringing together Japan, South Korea, Singapore, Israel, UAE, UK, and Australia to “build a secure, prosperous, and innovation-driven silicon supply chain from critical minerals and energy inputs to advanced manufacturing, semiconductors, AI infrastructure, and logistics.”

The project is led by Under Secretary of State for Economic Affairs Jacob Helberg with the explicit aim of reducing U.S. dependencies and protecting U.S. technology and infrastructure from “undue access or control by countries of concern.” In short, it aims to keep China out of U.S. and allied supply chains. The initiative aims to develop an insulated ecosystem spanning critical minerals, energy infrastructure, advanced manufacturing, and AI deployment within a network of core partners. Members are expected to build “trusted technology ecosystems” and coordinate responses to China’s “overcapacity and dumping”. The major difference from Sullivan’s strategic denial approach is President Trump’s willingness to permit sales of older generation chips to Chinese companies.

Both Sullivan’s export controls and Trump’s Pax Silica rest on the same premise: U.S. control of semiconductors and their supply chains creates dependencies for both partners and adversaries that give Washington long-term leverage. Sullivan believed widening the technology gap through export controls would keep allies locked into American technology and adversaries locked out.

Trump’s Pax Silica believes coordination with allies while maintaining tiered technological advantages (vetted partner access to newest generation technology) achieves similar structural outcomes—long-term dependence on U.S. innovation—while generating revenue for American firms and binding U.S. allies into long-term supply chain commitments. The core difference is that the Biden administration weaponized the Foreign Direct Product Rule to freeze China’s tech capabilities in place and restrict access to any U.S. technology. Trump monetized access for older chips with the aim of keeping Chinese companies dependent on older U.S. technology (securing profit for American companies) while restricting them from the newest U.S. innovation.

There is no clear “right” answer on which policy works best. Trump’s Pax Silica may have a slight edge in that it also addresses the other end of the supply chain—deliberate efforts to reduce U.S. dependence on China’s monopoly in rare earth minerals processing. But both Sullivan’s export controls and Trump’s Pax Silica share the same underlying assumption: that semiconductor advantages create lasting strategic leverage. This rests on treating chips as the binding constraint in technological competition. Chips remain the center of gravity for U.S. technology strategy, and that focus is justified. But semiconductor advantages mean little if the materials enabling every chip generation remain under Beijing’s control. Washington cannot win the chip race while losing the supply chain competition beneath it.

China’s mineral monopoly

China holds astounding advantages at the base of the technology supply chain and has actively sought to restrict U.S. access in retaliation to U.S. exportv controls. This matters because while the United States maintains leads in advanced semiconductor design and cutting-edge fabrication, the U.S. will have to focus both on innovating across the supply chain to preserve its existing lead in chip manufacturing with the continued release of new models as well as catching up to China’s mineral and rare earth monopoly.

According to the International Energy Agency, China is the leading refiner for 19 of 20 strategic minerals, controlling 70 percent average market share. This concentration has intensified in recent years despite pledges to diversify supply chains. China possesses near-monopoly conditions for materials like gallium (98 percent of global production) and germanium (68 percent). But the numbers understate the problem. China’s dominance extends across the supply chain from extraction to processing expertise, including accumulated industrial knowledge, workforce capability, and operational experience which, presently, no other country can match. Mineral experts I have spoken with argue these advantages will take years to be replicated in the West.

AI experts in Washington I interviewed see it differently. They argue U.S. critical mineral gaps are easier to bridge than China’s semiconductor gaps. Alvin Camba’s recent analysis in War on the Rocks offers a useful framework here. He explains that rare earth leverage “burns” while semiconductor leverage compounds. In other words, Beijing’s mineral restrictions are most potent at the moment of imposition, but they create a cadence of impacts which can have a jumpstarting effect. First, the price spikes, which then mobilize Western governments to unlock capital and accelerate projects which drive diversification.

But closing those gaps would still require sustained alignment of capital investment, regulatory reform, streamlined permit processes, research infrastructure, and long-term procurement guarantees. The prospects for that alignment in the United States seem grim following Trump’s targeting and defunding key universities critical for innovation in the sector and ongoing targeting of core allies necessary to support western reindustrialization of critical and rare earth element processing.

Meanwhile, China is increasingly confident in demonstrating its leverage over mineral supply chains. Beginning in mid-2023, Beijing implemented export restrictions on strategic minerals. In August 2023, China introduced licensing requirements for key rare earths, notably gallium and germanium. By December 2024, China banned all exports of these materials to the United States, along with antimony and superhard materials. In April 2025, China added seven medium and heavy rare earth elements—samarium, gadolinium, terbium, dysprosium, lutetium, scandium, and yttrium—to export control lists. By October 2025, China expanded restrictions to cover rare earth production equipment and related technologies.Following Trump-Xi discussions in late October 2025, China suspended many of these restrictions for one year beginning November 2025, though the core licensing system for rare earths implemented in April 2025 remains in force.

Beijing showed that it too can use strategic denial in its competition with Washington. Chinese officials created immediate pressure on Western mineral-heavy industries without fabricating a single advanced semiconductor. Then, Chinese authorities strategically dialed restrictions back as bargaining leverage. However, Beijing may have overplayed its hand. Camba argues correctly that China’s export controls accelerated international coordination through the Quad, the U.S.–Australia Critical Minerals Partnership, and the G7 Minerals Security Partnership to pool stockpiles, share data, and align finances. These help with bridging near term supply gaps, but they are not a long-term substitute a coherent strategy for rejumpstarting U.S. locally-based and friend-shored supply chains, as argued by Peter Harrell.

Refining capacity and supply chain control compound over decades of expertise, methods, and experience. These advantages are unlikely to be overcome quickly through investment alone (though this helps). The U.S. let China build this lead when it reduced its strategic stockpile and offshored materials processing decades ago.

This partly explains Beijing’s confidence. Deng Xiaoping famously said in the 1980s, “The Middle East has oil. China has rare earths.” China doesn’t necessarily need to match American chip design today to exercise leverage over global technology supply chains tomorrow. Beijing’s grip on mineral processing gives it options to restrict, delay, or reshape markets in ways that compound rather than depreciate. That is a tough reality with which multiple U.S. administrations have and will have to address. Maybe Pax Silica is the means of undercutting China’s monopoly, but maintaining a cohort of critical partners when Trump’s other policy priorities (like Greenland) risk ripping apart rather than deepening key U.S. partnerships.

How this impacts the GCC?

Pax Silica lays the foundation for a new semiconductor order, but Trump’s actual implementation reveals a deeply transactional logic beneath the institutional framework. The official structure suggests membership determines access—Pax Silica signatories get current-generation chips, non-members receive degraded technology or nothing at all. Yet Saudi Arabia receives advanced Blackwell chips despite not joining the coalition, while the UAE’s G42 maintains deep Chinese research ties even after formally divesting from Chinese firms to secure U.S. chip access. The pattern shows chip distribution operating through deal-making rather than exclusively via institutional commitment. This suggests that coalition membership is not the only condition for access, but also a nation’s willingness to pay, align politically, and accept conditions that preserve U.S. control over future upgrades. Through this view, Pax Silica provides the diplomatic and coordination umbrella for competing against China, but the underlying mechanism for determining chip access is still transactional. This creates ample space for members to hedge if Pax Silica does not deliver the value they hope long-term.

The UAE is a unique case. The UAE, for now, is the only GCC country apart of Pax Silica. The UAE had deep AI partnerships with Chinese firms in 2023, then U.S. policymakers used these ties to justify an attempted to block Emirati AI giant G42 from accessing U.S. advanced chips. Eventually, G42 agreed to divest from Chinese technology and severed ties with ByteDance and Huawei as a precondition for chip access. Another Abu Dhabi investment vehicle—Lunate—took over management of G42’s Chinese-focused fund.

The UAE’s calculus favors U.S. partnership for the time being because the U.S. has promised and slowly delivered on providing technology Beijing cannot match. While China prioritizes the autarkic development of its own AI ecosystem, the UAE is receiving chips and infrastructure far superior to what Chinese firms can deploy at home. The scale of this access is already substantial and growing. In November 2025, the Commerce Department approved shipment of approximately 35,000 high-end Nvidia chips to Saudi Arabia’s Humain and the UAE’s G42. The approval followed Saudi Crown Prince Mohammed bin Salman’s Washington visit, where the kingdom announced $1 trillion in new U.S. investments. More significantly, Gulf states are receiving Blackwell-generation chips—more advanced than the H200s approved for China. In raw silicon terms, the UAE is leapfrogging China entirely.

Reported plans dwarf even these initial approvals. The Trump administration is reportedly considering deals that would allow more than one million advanced AI chips to be exported to the UAE alone—enough to build one of the world’s largest AI training clusters. The UAE has announced construction of a five-gigawatt AI campus in Abu Dhabi, the largest AI infrastructure project globally, and OpenAI is weighing opening a data center there. In neighboring Saudi Arabia, Humain received commitments for 18,000 Blackwell chips in a first phase, with Amazon investing more than $5 billion alongside Humain to develop an “AI zone” with AWS infrastructure.

Yet chips alone do not fully make an AI ecosystem, and this edge only matters if the UAE can translate hardware access into indigenous capabilities. Here, Chinese companies may offer the UAE only a little help. As we explored in previous papers, Chinese authorities view the Gulf as customers and capital sources, rather than collaborators in frontier technology development. This could change now that the UAE (and Saudi Arabia) are set to received an influx of new chips. Still, framing China and the UAE as competitors oversimplifies a more complicated relationship. U.S. pressure on the UAE to divest from China is unlikely to untangle those ties. Both countries maintain deep ties at the personnel and institutional level, creating channels likely to outlast corporate partnerships.

The lesson for Pax Silica is that dependencies built on technology advantages remain contingent on maintaining those advantages. If U.S. terms become too restrictive, Chinese alternatives remain available. If China does not catch up to U.S. capabilities, the UAE will still be in an advantageous position. Though, the experience of other Pax Silica signatories will require further observation to assess how their individual experiences and competitive advantages shape their individual roles in the emerging AI ecosystem.

Delivering is a different story

The Trump administration has adopted what Alasdair Phillips-Robins describes as the “big announcement, slow follow-up“ tactic. He explains that after promising millions of chips to Saudi Arabia and the UAE in May 2025, it took six months for first approvals, with Commerce signing off on just 35,000 chips—far less than publicly discussed figures. The delay came as Commerce Secretary Howard Lutnick pushed Emirati leaders to follow through on their promises of U.S. investment before signing off on exports, with national security concerns also holding up approvals. If H200 shipments to China follow similar patterns, White House and Commerce officials will negotiate license conditions with Nvidia, Chinese buyers, and potentially the Chinese government, creating extended implementation timelines that allow for tactical adjustments.

The pull factor for GCC countries is also already delivering major wins for Gulf AI ecosystems. The Gulf states’ “triple advantage“—cheap electricity, massive desalination capacity for data center cooling, and strategic geography serving four billion users with sub-100-millisecond latency—creates structural pull for U.S. technology firms seeking to escape domestic power constraints and permitting bottlenecks. Projected U.S. AI data center demand of 130 gigawatts by 2030 far exceeds available domestic generation capacity under development. These power constraints make the Gulf’s advantages a potential structural necessity for U.S. AI ambitions which Washington cannot replicate domestically. Trump has chosen to leverage Gulf infrastructure rather than constrain AI development at home, but this may create some degree of mutual dependency. Washington may maintain control over chip access and pathways to new technology, but cannot escape the fact that computation increasingly migrates to facilities it does not control.

This muddies U.S. leverage. The United States can condition chip exports on Gulf states ending joint military exercises with China—both the UAE and Saudi Arabia have conducted these exercises in recent years. Washington can demand verifiable evidence that the UAE is complying with its demands for limiting Chinese access to critical U.S. technology. The Commerce Department can require enhanced documentation demonstrating legitimate end use, making Gulf states prove they are not serving as transshipment points for chips ultimately destined for China. But, the real question lies in how the U.S. responds if U.S. officials do not find the UAE is operating within U.S. requirements or how the UAE responds if U.S. domestic political constraints threaten to undermine the original terms of cooperation?

Yet the diversion risk remains a political minefield in Washington, and a number of experts I spoke with still question the decision to make GCC states core AI partners. The near-term concern is direct diversion; the longer-term worry involves more sophisticated technology arbitrage where Gulf states use advanced AI capabilities to provide services to Chinese entities, effectively laundering compute capacity without physically transferring chips.

The GCC’s commitment to trillions in AI spending built on the back of American chips is a strong enough economic incentive that Trump will take a risk and monitor compliance in real time.

How then should we view Trump’s semiconductor policy?

Finally, we get to Trump selling H200s to China. What does it mean that Trump can approve selling advanced chips to China for the right price? It means semiconductor advantages work differently than most strategic assets. Trump can negotiate chip access deal by deal. He cannot, however, negotiate away mineral dependencies that took decades to build. The Administration’s frantic efforts to secure critical mineral stockpiles—threatening Greenland invasions, leveraging Ukraine and DRC peace negotiations for resource access, rushing new refining plants—reflects recognition of this asymmetry. Yet rash policymaking cannot solve structural impediments. Industry experts assess U.S. efforts to rebuild advanced processing capacity will require ten to fifteen years, assuming sustained political commitment across election cycles, comprehensive regulatory reform, and long-term procurement guarantees—conditions far from assured.

China’s mineral dominance operates by different rules entirely. Beijing’s lead in processing and refining has compounded over decades and continues to appreciate. Chinese strategists accept near-term chip performance gaps to maintain long-term control over materials enabling all chip generations. When Trump offered H200 chips, Beijing appeared to reject them, with officials indicating they “want semiconductor independence” and prefer to “prop up Huawei.” From China’s view, mineral dominance provides leverage over the entire technology stack because every semiconductor, battery, and data center requires processed material inputs. Chip shortages force designers to find alternative suppliers or optimize software. Missing rare earths or specialized alloys stop production lines entirely, a harsh lesson the U.S. industries learned during COVID-19.

Both Biden and Trump treated semiconductors as strategic assets comparable to oil in the twentieth century. Control over oil supplies has shaped international order for generations. But what if semiconductors behave more like steel in the nineteenth century? Steel was a critical input, but its technological frontier advanced and capabilities diffused to new countries, which created new centers of innovation and expertise. Britain dominated steel production in 1850. By 1900, American and German mills had surpassed it through newer processes and scale of production. The advantage for the UK proved temporary because steel production depended on replicable industrial techniques. Semiconductors could follow this pattern, though the technology necessary to build a foundry and manufacture chips is still tightly regulated, which gives the U.S. advantage. Each new generation of chip creates opportunities for competitors to leapfrog through new innovation, process optimization, or simply outspending incumbents. Lasting strategic advantages require generational commitment to build and offer no shortcuts to replicate. They are rooted in accumulated operational knowledge and resource control that competitors cannot easily engineer around.

But, the same story could also be true for China’s lead in critical mineral refining. One major investor in the mining sector I interviewed argued that once supply chains realign and processing and refining technology diffuses to other parts of the world, the mineral race will slow and so will dependence on China. China may hold the monopoly now, but, like steel, new centers of innovation could emerge and undercut that dominance. This is perhaps the emergence of the economic dimensions of multipolarity as global supply chains adjust to reduce reliance on current problematic chokepoints, wherever they may exist.

Washington has to both manage its chip lead and bridge the mineral gap. That is not a small feat. Pax Silica may succeed on its own terms, but Trump’s transactional technology policies also have downsides. It raises the question: if chips are a source of American power, and that dominance can be accessible to other countries for a price, then semiconductor access operates as a service to be purchased, not a strategic asset to be protected. China’s rejection of H200 chips suggests Beijing may wholly reject the premise of any form of dependency on western technology moving forward. There is still an ongoing debate over how far Beijing should lean into autarky, but there is a general consensus that dependency on the west must end. What does breaking it look like? For now, China is accepting performance gaps today to maintain structural control over critical minerals tomorrow with a belief that they will eventually overcome the performance gaps.

Thank you for reading. Please like and subscribe for more essays.

Subscribe now

This discussion featured Margaret Myers and Tong Zhao, both who bring unique and deeply informed viewpoints on Chinese foreign policy on Latin America after the fall of Maduro and what could be a potential domino effect in the region.

In the discussion, our panelists go deep into what this means for China’s foreign policy toward Latin America, Taiwan, and more.

This is conversation you will not want to miss.

Subscribe now

The Trump administration has thrown international affairs into new levels of volatility. The chaos which has ensued has created a very difficult world to navigate for many countries. China is not immune it’s impact. The clear lesson many are finally realizing is that Trump operates on instinct, domestic politics, and personal grievances rather than an identifiable coherent strategy. That is not to say people (including the National Security Counci) have tried to make a Trump doctrine coherent and we have to acknowledge that there are trends which show Trump’s style of governance. He launches sudden military strikes, announces sweeping tariffs out of nowhere, reverses diplomatic positions overnight, and issues threats he may never follow through on. For the many reading this who have overheated their iphones doom scrolling on Venezuela, Greenland, and Iran (to name a few), you are not alone. Think about the Chinese strategist trying to interpret this these signals? They have to manage Beijing’s responses to American foreign policy while trying to decode a decision-making process and doctrine that nobody, including Trump’s own advisors, seems able to predict.

Subscribe now

This unpredictability hands China some real opportunities. America’s traditional partners are experiencing major whiplash, especially in 2026. European allies, Middle Eastern governments, and Latin American states that built their foreign policies around American predictability now watch commitments evaporate and threats materialize—or not—without any clear pattern. China will almost certainly attempt step into this void and try to sell what Washington presently won’t: consistent engagement, clear terms, and foreign policy that doesn’t depend on presidential mood swings. Will China take advantage of its positionality with some partners? Likely yes, which means they may aim to be extractive, but the selling point is predictability. Countries exhausted by Trump’s chaos become more receptive to Chinese alternatives, particularly when Beijing promises infrastructure investment and trade relationships that won’t disappear because of a 3 a.m. Truth Social post. Point and case: Trump’s persistent strong arming of Canada has pushed them closer to Beijing. Canada’s prime minister visited Beijing this week hoping to secure deeper ties with China, which will not land well with Washington. In a statement, Canada’s PM stated that growing ties with Beijing would help position China and Canada “for the new world order".

But China’s opportunity rests on a fatal assumption: that China can actually read Trump and avoid the traps he’s setting. The evidence says they can’t. Trump’s recent moves defy traditional strategic analysis. He threatened Iran with military force if authorities killed protesters, then sat back as casualties surged past 1,850 deaths. He pivoted to diplomatic talks, then canceled everything, then threatened “very strong action” over scheduled executions. Then, Trump spent months conducting incremental strikes on Venezuelan boats, then suddenly launched a full decapitation operation grabbing Maduro. He hits Iran’s trading partners with sweeping secondary 25 percent sanctions—which catches China, Iran’s biggest customer—but provides zero implementation details. Beijing has to guess whether Trump means it or whether he’s just creating negotiating leverage.

Trump wants adversaries off-balance, and chaos has been his most successful avenue for disadvantaging those he sees as competitors. For China, this turns every response into a potential disaster. Push too hard supporting Iran, and Trump might actually enforce those tariffs—or worse. Pull back too quickly may show weakness that invites more pressure. China typically calibrates responses based on clear (or slightly fuzzy) understanding of American red lines. That approach collapses when the red lines themselves shift constantly and deliberately stay blurry.

China faces a deeper challenge. Its partners keep collapsing, and each collapse scrambles the strategic picture. Venezuela, where China holds massive debt and oil interests, just saw American special forces capture its president. Trump now says the U.S. will “run the country.” Iran, another crucial Belt and Road partner and oil supplier, teeters between regime collapse and American military intervention—maybe both?. Syria’s Assad regime, which China defended in the UN Security Council for years, already fell. Each disaster forces Beijing to reassess its tactical positions and its fundamental assumptions about American willingness to use choas and force. Chinese strategists and policymakers end up frozen, spending so much time analyzing what might happen that they can’t respond decisively to what actually happens.

Trump's 2025 National Security Strategy made this objective explicit through the "Trump Corollary" to the Monroe Doctrine, which declares the U.S. will "deny non-Hemispheric competitors the ability to position forces or other threatening capabilities, or to own or control strategically vital assets, in our Hemisphere." Driving China from Latin America and Greenland—where Trump has warned China may be seeking access—is at least part of the motivation behind the current chaos as he reasserts what he views as America's natural sphere of influence. China could push back against this pressure, but doing so would be like walking into an unfamiliar dark room with a floor covered in sharp tacks and a big man with a bat hiding somewhere in the shadows, hoping to emerge on the other side in better shape than when you entered.

Chinese decision-makers may now burn enormous resources trying to parse American intentions, and this slows everything down. When Trump threatens Iran, Chinese analysts face impossible questions: Is he serious? Will he actually strike? When? How? Where should we position ourselves?

These questions have no answers, but the time spent debating them delays China’s responses and creates the kind of hesitation that undermines Beijing’s carefully cultivated image of competent certainty. The analytical burden applies everywhere—Venezuela, Greenland, Taiwan, trade policy. Every issue demands extensive scenario planning for Trump moves that may never happen or may happen in completely unexpected ways.

China will keep exploiting tactical openings. Greenland’s resources and Arctic position offer ways into European influence if Trump’s aggressive pursuit alienates Denmark. Venezuela’s chaos creates opportunities to extract concessions from whoever ends up governing. Iran’s desperation for economic partners makes it more dependent on Chinese trade despite American sanctions. But these tactical moves happen within a strategic environment of profound instability. The macro-level question—how does China actually respond to American power projection under Trump—has no stable answer.

The real challenge is that Trump mixes genuine unpredictability with calculated ambiguity, personal vendettas with strategic goals, domestic politics with foreign policy in ways that resist systematic analysis. There are patters, but as soon as a pattern emerges, it is broken. China built its system of foreign policy analysis on careful long-term planning and meticulous risk assessment. Trump’s approach creates a minefield where every step forward carries unknown dangers. It may force China into reactive postures and create hesitation, though it is certainly unlikely to slow its global ambitions.

Whether this represents deliberate genius or reckless improvisation remains unclear. And that uncertainty is exactly what makes it so hard for Beijing to counter. When your opponent doesn’t follow any discernible rules, you can’t develop a winning strategy—you can only react and hope you guessed right.

The question though is when does Trump’s improvisation and unpredictablility alienate core U.S. partners to the point of distancing ties to avoid backlash?

Subscribe now

Subscribe now

If China sees the GCC as a deployment market for its own AI technologies, the United States, under the Trump Administration, has made the GCC a critical node in its AI ecosystem. That positionality comes with strings attached for GCC countries who see their path to near-to-mid term AI development through Washington and Silicon Valley. However, a major question remains: what forms of leverage and comparative advantages do the Arab Gulf states bring to the table, which give it bargaining power between the U.S. and China?

To answer this question, we must start with the GCC’s comparative advantage.

Gulf states center their AI infrastructure advantage on converting hydrocarbon wealth into computational power—what analysts term “petro-compute.” Abdullah Alzabin identifies what he calls a “triple advantage”: Saudi Arabia and the United Arab Emirates enjoy some of the world’s lowest electricity costs—unsubsidized power at $0.10/kWh versus $0.17 in the U.S.—driven by abundant natural gas and rapidly expanding solar generation. Their desalination systems produce roughly 40 percent of global desalinated water output, providing the cooling capacity that power-intensive data centers require. Geographic location matters too: Gulf states can serve four billion users across Africa, Central Asia, and the Middle East with sub-100-millisecond latency. Add nearly $5 trillion in sovereign wealth fund assets, and Gulf states possess the resources to build AI infrastructure at speeds and scales that would strain energy and water systems elsewhere. Yet whether these advantages generate genuine technological leverage or simply position Gulf states as well-capitalized landlords for technologies developed elsewhere remains sharply contested.

The distinction between AI training and inference divides analysts assessing Gulf leverage. Training new models requires advanced semiconductors—chips subject to U.S. export controls that Washington extended to Gulf states in October 2023. Running those models to answer user queries (inference) needs less sophisticated chips but consumes enormous power as billions of users make trillions of requests daily. Optimists see an opening: Gulf states can dominate inference through cheap energy even if denied access to cutting-edge training chips. Skeptics respond that hosting inference operations without developing the underlying models leaves Gulf states as well-capitalized landlords, not technology creators. Those who argue the Gulf is employing a hedging strategy observe a third option: the training-inference split creates optionality to play U.S. and Chinese technology providers against each other, securing better terms from both.

Writing in Foreign Affairs in December 2024, Daniel Benaim—former Deputy Assistant Secretary of State for Arabian Peninsula Affairs in the Biden administration—articulates an optimistic “diffusionist” view of Gulf petro-compute capacity. Benaim argues that Gulf states, “armed with chips, sovereign wealth, and abundant energy,” are positioned to “surpass Europe and India in terms of AI infrastructure—eventually becoming the world’s third biggest hub for AI computing power, behind the United States and China.” In his framing, “computing power has now taken its place alongside crude oil as a pillar of the U.S.-Gulf relationship,” transforming Gulf states into essential partners for U.S. AI firms seeking to escape domestic power and permitting bottlenecks. Benaim emphasizes that Gulf connectivity to Africa, Central Asia, and the Middle East could extend the reach of the U.S. “AI stack” to billions of users, while displacing China as the Gulf’s primary technology partner—”a big win for Washington over Beijing.”

In this paper’s reading, Benaim’s framing—while emphasizing U.S.-Gulf strategic alignment and mutual benefits—implies genuine Gulf agency and leverage in U.S.-China technological competition. The UAE, according to Benaim, now has “the highest rate of AI adoption of any country,” while Saudi Arabia has invested billions in U.S. tech firms since 2016 and integrated AI into flagship ventures. Gulf regulatory flexibility, cheap energy, and massive capital pools create structural advantages that make partnership attractive—even compelling—for both American and Chinese technology providers.

Abdullah Alzabin develops the most comprehensive case for Gulf strategic opportunity in AI infrastructure. Alzubin argues that U.S. domestic power constraints—projected 130 GW of additional AI data center demand by 2030 against only 30 GW of new U.S. gas generation capacity—will force American policymakers to “prioritize power capacity for strategically vital AI model training while moving inference operations, the less critical task of running these models, abroad.” This creates an opening for GCC states to become “major suppliers of AI inference through gigawatt-scale computing infrastructure,” leveraging what Alzubin terms a “triple advantage” of power infrastructure, strategic geography, and financial capability. Unlike training, he notes, “inference computing requires less advanced semiconductors—which face fewer export restrictions—enabling the GCC to build substantial capacity even under current U.S. export controls.” Alzubin frames this as a potential transformation analogous to Kuwait’s 1938 oil discovery—a moment when “centralized planning and execution” combined with structural advantages could enable GCC states to “embed themselves in an industry that could transform their economies and build vital complexity.” He argues that “the cumulative power needs from billions, and eventually trillions, of daily inference operations will exceed training requirements” as AI evolves toward “inference-time reasoning—where models perform complex calculations while serving users rather than at training.”

Yet even Alzubin acknowledges critical value capture risks, warning that GCC states could end up “investing in infrastructure that, while necessary, lacks differentiation—a potentially commoditised asset with high fixed costs and limited pricing power.” He points to the telecom sector’s cautionary tale: AT&T’s exclusive iPhone partnership in 2007 saw the carrier valued at $250 billion to Apple’s $105 billion; today “AT&T’s market capitalisation stands at $165 billion, while Apple has reached $3.5 trillion—demonstrating how value can accrue to higher layers in the technology stack.”

Yet other analysts frame petro-compute capacity more skeptically, viewing it less as a path to technological sovereignty than as a new dependency relationship. Layla Ali, in a broader analysis of Gulf states as unconventional middle powers with rising strategic autonomy, warns that “much of the Gulf’s AI ecosystem relies on imported technologies, foreign-owned platforms, and expatriate expertise. This dependency raises questions about technological sovereignty and the long-term sustainability of innovation ecosystems.” While Ali sees Gulf states successfully exercising agency through multi-vector diplomacy and norm-setting, she flags technological reliance as a key vulnerability that may constrain these ambitions. Her assessment is blunt: “Without a strong domestic base in research and development, the region risks becoming a technologically advanced consumer but not a true producer of frontier technologies.”

Mohammad Rashed Albousa, et al writing in Nature Humanities & Social Sciences Communications, reinforce these concerns in their empirical analysis of Gulf workforce preparedness. Examining data-center and cloud investments across all six GCC states, Albousa and colleagues find that large-scale capital deployment has not produced a commensurate rise in locally owned innovation outputs, particularly where highly skilled technical roles are dominated by expatriates. Their study asks whether “investments in compute infrastructure [are] matched by an equally robust build-out of skills, incentives, and governance,” finding gaps between infrastructure spending and indigenous capability development. This empirical pattern—combined with Layla Ali’s strategic-autonomy concerns and Alzabin’s value-capture warnings—suggests Gulf states risk trading oil dependence for chip and cloud dependence, becoming infrastructure providers while core intellectual property, algorithms, and technical talent remain external.

The fact that Gulf cooperation attracts both U.S. and Chinese technology firms—the very dynamic Benaim celebrates—may not be a reflection of Gulf leverage but rather the Gulf’s structural position as a well-capitalized deployment market rather than a source of indigenous innovation. Gulf states must still import advanced chips, whether from the United States (as Benaim advocates) or potentially from China; external providers develop cloud architectures and AI platforms that Gulf firms adapt locally; and the U.S. can withhold semiconductor shipments if Gulf states “violate terms” (as Benaim acknowledges), underscoring ongoing dependency rather than genuine autonomy. Alzubin himself identifies geopolitical vulnerability as a critical risk: “Expanding U.S. export controls to less advanced semiconductors threaten to curtail GCC ambitions,” while technological dependence “mirroring their reliance on U.S. military equipment—could further constrain diplomatic flexibility.”

A third perspective treats Gulf petro-compute as deliberate hedging between Washington and Beijing. Sam Winter-Lev argues that “although the Gulf states are eager for advanced AI chips that for now only the United States can provide, they also have strong and enduring incentives to hedge their bets, playing the major powers off against each other to extract concessions.” He notes that “Saudi Arabia and the UAE both have powerful incentives to hedge their bets, given American domestic political instability and the enduring, if eternally frustrated, U.S. desire to ‘pivot’ to Asia,” while China remains Saudi Arabia’s largest oil customer and trading partner and the UAE’s top non-oil trading partner.

From this perspective, the UAE’s willingness to “rip out and replace its Huawei hardware” (as Benaim notes occurred after 2023 U.S. chip export restrictions) represents tactical positioning rather than strategic commitment to either camp. The UAE’s trajectory illustrates this hedging in practice: deep AI partnerships with Chinese firms through 2023, followed by a sharp pivot to U.S. alignment through the Microsoft-G42 deal, which the U.S. Commerce Department’s Bureau of Industry and Security conditioned on G42 severing ties with ByteDance and Huawei. Yet even as G42 divested from Chinese firms, Winter-Levy notes, “a new Abu Dhabi investment vehicle has taken over the management of G42’s Chinese-focused fund, and, like G42, the new vehicle is overseen by the Emirati national security adviser.” Rather than signaling weakness, this maneuverability demonstrates precisely the optionality petro-compute creates: if U.S. terms become too restrictive or enforcement too intrusive, Chinese alternatives remain available; if China cannot match U.S. capabilities, American partnerships deliver superior technology. Petro-compute, in this framing, generates bargaining power to credibly threaten alternative partnerships in negotiations with both Washington and Beijing.

The competing views among Western analysts, however, may obscure a more fundamental dynamic visible in Chinese assessments. From a Chinese analytical perspective, Gulf petro-compute ambitions—however grounded in genuine structural advantages in energy, water, geography, and capital—reinforce rather than challenge the complementarity framework. Chinese sources consistently note that Gulf investments in AI and semiconductors—however massive—do not translate into technological self-sufficiency but rather create dependencies on external providers.

Gulf capital finances the infrastructure through which Chinese (and Western) technology achieves global reach, but the core capabilities—chip design, AI algorithms, systems integration—are only accessible from external markets. This assessment aligns closely with Layla Ali’s conclusion that Gulf states risk becoming “consumers but not producers” of frontier technologies, regardless of the scale of infrastructure investment—a concern that persists even alongside the indigenous model development and the strategic positioning opportunities that Alzubin identifies.

Whether the Arab Gulf states partner primarily with U.S. or Chinese firms, the underlying dynamic remains constant in Chinese analysis. Gulf resources, in China’s view, enable others’ technologies to scale, but do not generate indigenous capabilities that shift the relationship from complementarity to genuine interdependence. This makes petro-compute capacity valuable to China as a scaling partner and alternative deployment zone, but not as a technological peer. Chinese analysts would likely interpret the urgency of both Benaim’s and Alzubin’s arguments—and the Trump administration’s willingness to export advanced semiconductors despite “chip hawk” concerns about leakage—as validating their assessment that great powers compete for Gulf capital and markets precisely because Gulf states control resources but not underlying technology. The Gulf’s ambition to become “the world’s third biggest hub for AI computing power” (as Benaim writes) or to leverage AI inference as what Alzubin terms “compute as the new currency for power” does not, in Chinese framing, translate into becoming the world’s third source of AI innovation. Instead, it positions Gulf states as a strategic arena where Chinese and American technologies compete for deployment, investment, and political alignment—a role that is valuable but fundamentally different from the technological self-sufficiency Gulf diversification strategies officially pursue.

The Leverage Question

The evidence suggests that petro-compute generates tactical bargaining power rather than strategic leverage. Gulf states demonstrably possess the ability to extract concessions from both Washington and Beijing—the UAE’s pivot from Huawei to Microsoft while maintaining Chinese partnerships through alternative investment vehicles proves this maneuverability is real, not rhetorical. Yet this optionality operates entirely within parameters set by external technology providers. The urgency with which both American and Chinese firms compete for Gulf capital and deployment markets reveals the Gulf’s value proposition: not as innovation partners but as well-capitalized infrastructure providers. When Benaim celebrates Gulf states becoming “the world’s third biggest hub for AI computing power,” he inadvertently highlights the limitation—third in infrastructure deployment, not third in technological capability. The Gulf can credibly threaten to switch partners, but cannot credibly threaten to develop indigenous alternatives that would render either partner dispensable. This is bargaining power, certainly, but it is bargaining power that evaporates the moment Gulf states attempt to move beyond complementarity toward genuine technological interdependence.

The Chinese analytical perspective—that Gulf investments reinforce rather than challenge dependency relationships—may prove more predictive than the optimistic diffusionist view or even the hedging framework. Whether Gulf states partner with U.S. or Chinese firms, the fundamental dynamic persists: external providers control chip design, AI algorithms, and systems integration while Gulf capital finances the infrastructure through which those technologies scale globally. Alzubin’s telecom analogy is instructive precisely because it captures how value accrues to creators rather than hosts—AT&T’s exclusive iPhone partnership generated infrastructure revenue but Apple captured the transformative economic gains. Yet the diagnosis does not exclude Gulf agency, but it underscores the challenges to it.

For Gulf states pursuing genuine technological sovereignty as part of economic diversification, petro-compute (like hydrocarbons) creates optionality. Infrastructure advantages generate bargaining power to negotiate technology transfer, attract research partnerships, and fund indigenous capability development in ways that pure capital deployment could not. The leverage question, then, is not whether petro-compute advantages are real—they demonstrably are—but how Gulf states choose to deploy them. This remains the leading long-term challenge and opportunity for GCC states.

Strategic choices matter: using cheap energy merely to host others’ inference operations would indeed create a more sophisticated form of resource dependence where computing infrastructure replaces crude oil as the Gulf’s primary export commodity. Yet deliberately leveraging infrastructure advantages to incentivize technology transfer, build research institutions, develop indigenous models, and cultivate local expertise could convert deployment capacity into genuine technological depth. Petro-compute advantages are tools, not outcomes; whether they generate strategic autonomy depends on whether Gulf states use infrastructure leverage to climb the value chain or accept the role of well-capitalized landlord.

Pax Silica and the Gulf’s Strategic Dilemma

The December 2025 launch of the U.S. State Department’s Pax Silica initiative—bringing together Japan, South Korea, Singapore, the Netherlands, the United Kingdom, Israel, the UAE, and Australia to build a “secure, prosperous, and innovation-driven silicon supply chain“—crystallizes the strategic tensions outlined in this paper. The UAE’s participation validates optimists’ claims that petro-compute generates genuine strategic value: Washington explicitly prioritizes “energy grids and power generation” and “compute” alongside semiconductors and recognizes that AI scaling requires Gulf energy advantages. Yet Pax Silica’s structure reinforces skeptics’ dependency concerns. Pax Silica focuses on semiconductor manufacturing as the coalition’s nucleus and is set to expand to countries contributing minerals and other inputs. This risks positioning the GCC precisely where Chinese analysts already place it: as resource provider rather than technology innovator. The initiative centers countries that control critical nodes in the semiconductor supply chain—advanced lithography equipment, materials production, packaging capabilities, and chip design—while the UAE contributes energy and infrastructure: the deployment layer, not the technology layer.

More critically, Pax Silica formalizes the “trusted partners” framework that could limit Gulf hedging. The initiative explicitly requires participants to protect “sensitive technologies” from “countries of concern” and build “trusted ecosystems,” requiring that members accept limitations on alternative partnerships. The U.S. conditions on the 2024 G42-Microsoft deal (severing Chinese ties, submitting to U.S. export controls) previewed what Pax Silica may now systematize in more rigorous form. This transforms GCC hedging from strategic optionality into tactical maneuvering within increasingly narrow bounds. Yet the UAE’s inclusion also confirms that petro-compute advantages matter enough to earn a place at the table of the flagship U.S. economic security architecture. It will be interesting to see if Saudi Arabia will join at a later time. The remaining question is whether Pax Silica enables genuine technological advancement for Gulf states to secure technology transfer and the development of its core AI capabilities or whether it formalizes complementarity as permanent architecture. The answer depends on whether Gulf states leverage infrastructure advantages to climb the value chain or accept well-capitalized landlord status in exchange for semiconductor access and coalition membership.

This is the next installment in a paper series on US-China competition over GCC AI ecosystems. While I give summaries of each of these papers, you should go and read each of the papers I cite in this piece.

This piece is the third in a series of deep dives into United States-China competition in Middle East AI ecosystems. Last week’s paper examined how Chinese scholars assess China’s self-sufficiency prospects under semiconductor constraints. This piece builds from that and provides a selective survey of Chinese voices on China’s AI cooperation with the GCC.

Subscribe now

I have been having a few meetings with scholars and policy researchers in the Washington tech bubble focused on new U.S.-GCC tech investments and GCC posturing as they develop their AI ecosystems. I realized that it was not immediately clear how Chinese scholars viewed its tech relationship with GCC countries. The more I researched, the more I believe the view from China on tech cooperation with GCC is underrepresented in the discourse and deserves more attention.

Beijing approaches the GCC from a position of technological self-sufficiency. When Chinese scholars assess AI cooperation with the Gulf Cooperation Council (GCC) states, there is consensus among Chinese scholars that the partnership as strategically valuable but technologically unnecessary. Chinese scholars tend to frame the tech relationship with the GCC as one of complementarity (互补, hùbǔ) rather than interdependence (相互依存, xiānghù yīcún). They view the GCC as a valuable partner for deployment and scaling, but that the flow of technology is a one-way street—from China to the GCC, not the other way around. In other words, China supplies AI infrastructure to build the Gulf’s AI ecosystem, and Gulf countries provide the capital and markets for China to fund the deployment.

This framing is consistent with how Beijing reckons with its position toward global AI competition. The Biden-era tightening of U.S. chip restrictions forced China to move to reduce its dependence on all foreign dependencies in its tech stack. Chinese scholars generally agree on the merits of technological self-sufficiency (自主可控, zìzhǔ kěkòng), but diverge on Beijing’s continued integration in global technology supply chains.

China Leads, Gulf Finances

Sino–GCC AI and technology cooperation, in its current period, can be understood through “complementarity over interdependence or co-innovation.” Under this framework, China and the Gulf states occupy structurally distinct yet mutually reinforcing positions. China functions as the primary supplier of advanced technologies and development pathways, while Gulf partners contribute capital, deployment environments, and political alignment on governance norms. The relationship is complementary in function but asymmetric in dependency.

This asymmetry is deliberate. Chinese scholars consistently avoid framing Gulf cooperation as technologically interdependent or indispensable to China’s domestic innovation ecosystem. Instead, the Gulf is positioned as a downstream arena for technology deployment, market expansion, and regulatory alignment. China’s interest lies not in absorbing Gulf-generated innovation into its domestic AI ecosystem (even though GCC states aspire to develop their own domestic AI industries), but in exporting mature or near-mature technologies abroad to scale usage, shape norms, and reduce third-party reliance on Western systems. For Beijing, cooperation reinforces China’s technological autonomy rather than diluting it through mutual dependence.

Fudan University scholar Sun Degang’s co-authored work with Wu Tongyu exemplifies this logic. Writing in February 2021, they describe China–Arab technology cooperation as a brand-new field characterized by functional complementarity: Gulf Arab states provide capital and deployment capacity, while China supplies technology, expertise, and development models. Crucially, this cooperation is framed as a pathway for developing countries to escape Western technological monopolies. The asymmetry preserves China’s technological sovereignty while extending its influence abroad. In an earlier piece, Sun and co-author Yahia Zoubir identified this pattern of asymmetric partnership as central to Beijing’s Middle East strategy—economic engagement that advances China’s strategic interests while avoiding the security entanglements that characterized Cold War-era great power competition in the region.

Technology cooperation is cast as an instrument of order-shaping, not mutual vulnerability. By extending Chinese systems abroad without entangling its own innovation pipeline, Beijing preserves technological sovereignty while expanding influence.

This framing is reinforced by Xiao Qian at Tsinghua University’s Center for International Security and Strategy. Xiao situated China’s AI cooperation with the Global South—including the GCC—within a broader critique of U.S. technology restrictions and small yard, high fence policies. In her view, China’s role is that of an enabler: providing technology, standards, and governance frameworks that help developing countries close AI capability gaps while resisting Western exclusionary practices. Yet this model presumes that China’s domestic AI ecosystem remains self-sufficient. The Gulf is valuable precisely because it does not substitute for, or interfere with, China’s internal technological base.

Asymmetric strategic complementarity

The relationship reflects what I refer to as asymmetric strategic complementarity. This describes a cooperative relationship in which two states’ capabilities align functionally but unequally, allowing the stronger actor to extend influence and scale its systems while preserving autonomy and avoiding reciprocal dependence. This helps explain a persistent tension in Chinese analyses. The Gulf is often portrayed as strategically important but never structurally necessary. This explains why Chinese scholars emphasize cooperation in deployment-heavy sectors—cloud computing, smart cities, data centers, AI applications—while remaining largely silent on joint frontier research or upstream semiconductor co-development. Much of this happens in China. The GCC is treated as a governance and scaling partner rather than a source of innovation critical to China’s technological trajectory.

This framework also illuminates a blind spot in Chinese scholarship. By emphasizing complementarity over interdependence, Chinese analysts tend to under-theorize Gulf agency—particularly the constraints imposed by security alliances and Western regulatory ecosystems. Yet as Tin El Kadi’s research demonstrates, GCC states seek to actively shape these technology partnerships rather than serving as passive recipients of Chinese systems. Local regulatory requirements, data sovereignty mandates, and strategic hedging between U.S. and Chinese technology ecosystems all reflect Gulf agency in determining the scope and limits of cooperation. The GCC’s renewed integration with Western technology stacks in late 2025 demonstrates that deployment arenas remain politically contingent.

Chinese Investment as Complementary

Is the “complimentary without interdependence” thesis reinforced by practical investments? The empirical record of Chinese technology investments in Gulf markets between 2022 and 2025 appears to support the assessment. Chinese firms have embedded their technological capabilities across cloud infrastructure, AI platforms, and smart-city systems in Gulf markets. While Gulf investment in China exists, it remains marginal compared to Chinese incoming investment flows.

In 2022, Alibaba Cloud and Saudi Telecom (STC) launched the Saudi Cloud Computing Company, opening two public-cloud data centers in Riyadh to serve domestic government and enterprise demand. The joint venture structure paired Alibaba’s cloud architecture and management stack with STC’s local network access and market position. Huawei Cloud followed in 2023, announcing a USD 400 million, five-year commitment to establish its first Middle East cloud region in Riyadh, delivering localized AI and cloud services to Saudi government and enterprise users. Huawei executives explicitly framed the project as a bridge for bringing additional Chinese firms and technologies into the Saudi market.

This outward-scaling logic intensified in 2025. At the LEAP technology conference, Tencent Cloud announced a USD 150 million Middle East cloud hub in Riyadh—its first in the region—designed to provide cloud and AI services for gaming, smart-city analytics, and enterprise applications aligned with Saudi Vision 2030. Each deployment follows the same model: Chinese cloud architectures, developed and proven domestically, are adapted for Gulf regulatory requirements and demand profiles, with Gulf capital enabling regional buildout.

AI-specific cooperation reinforces the pattern. In 2022, Chinese AI firm SenseTime and the Saudi Company for Artificial Intelligence (SCAI) formed SenseTime MEA, backed by SAR 776 million (approximately USD 207 million) in Saudi investment to deploy SenseTime’s AI technologies in smart cities, healthcare, and education. While the venture includes local development activity, the core models and intellectual property remain Chinese. Saudi capital enables localization and commercial scale, but it does not generate upstream innovation feeding back into China’s domestic AI ecosystem—this is tailored deployment of technology specifically for the Saudi market.

In the United Arab Emirates, Alibaba Cloud opened its second Dubai data center in 2025, expanding regional capacity to support finance, logistics, and digital-government clients under UAE data-localization rules. The expansion was explicitly tied to Alibaba’s global cloud growth strategy. Chinese firms also exported turnkey smart-city systems across the GCC. In Qatar, Huawei was contracted to deliver the smart-city digital backbone for Gewan Island, integrating citywide connectivity, command-and-control platforms, and on-site data-center infrastructure. In Kuwait, state-owned China Gezhouba Group secured USD 1.12 billion in contracts to build South Saad Al-Abdullah, the Gulf’s first large-scale smart and green city, exporting Chinese construction, digital-infrastructure, and urban-management systems at scale.

Across these cases, the investment profile consistently points toward embedding technology and expertise in the same pattern: Chinese firms deliver cloud architectures, AI platforms, and smart-city systems developed domestically, and Gulf partners contribute capital, regulatory accommodation, and demand linked to national development visions. Where localization occurs—such as Arabic-language AI applications or compliance with data-sovereignty rules—it remains at the application and deployment layer, not at the level of core AI research, semiconductor design, or foundational model development. These adaptations expand Chinese technology’s market reach. This pattern reflects what analysts have termed reverse data dependency: Gulf countries use Chinese technology platforms, but the algorithms process locally generated data, enhancing Chinese firms’ algorithmic capabilities through access to diverse regional datasets without requiring fundamental technology transfer from the Gulf to China.

Identifying the GCC’s Function

Chinese investment and record of cooperation supports the complementary, not interdependent hypothesis. China does not treat the GCC as a critical upstream partner for its own AI production or innovation ecosystem. Instead, the Gulf serves three interlinked strategic functions: 1) it is a proving ground for Chinese technology systems outside Western markets, 2) a source of investment capital that accelerates Chinese firms’ global expansion, and 3) a governance partner helping Beijing build alternative technology frameworks beyond U.S.-dominated institutions. China does not need Gulf expertise to sustain its AI trajectory—but the trend in investments suggests it wants Gulf markets, capital, and political alignment to extend the reach of its technological systems.

Yet this relationship unfolds within a competitive GCC landscape. Roa Al Shidhani and Saranjam Baig underscore that economic disparities among GCC states generate competition for Chinese technology investments, particularly between Saudi Arabia, the UAE, and Qatar. While all six states pursue economic diversification through digital transformation, differences in hydrocarbon endowments and wealth levels shape their respective approaches. Saudi Arabia and the UAE, with their larger resource bases and GDP per capita, compete most directly for flagship Chinese AI and cloud partnerships, while smaller GCC economies seek more targeted engagements. This intra-GCC competition strengthens China’s bargaining position, allowing Chinese firms to negotiate favorable terms as Gulf states vie to position themselves as regional technology hubs.

The result is a relationship that is stable and limited, with possibility to scale if geopolitical dynamics change. It is resilient so long as political conditions between China and the GCC allow deployment, yet limited in its capacity to transform the underlying balance of technological power. So far, those political conditions are bending in favor of stronger US-GCC technological cooperation. China can walk away without structural damage; the Gulf cannot. Understanding this asymmetry is essential for GCC policymakers seeking to navigate between access and autonomy—and for analysts assessing how multipolar competition in AI is likely to evolve. Both proponents of greater Chinese technological self-reliance and advocates of continued openness to global markets agree on one fundamental point: the GCC provides valuable markets, capital, and strategic alignment, but not the technological inputs essential to China’s AI development. For China, Gulf cooperation offers strategic benefits without creating structural dependencies. For the Gulf, the challenge remains converting market access into genuine technological independence.

Complementarity may be sufficient for China’s strategic ambitions, but for the Gulf, the harder question remains unanswered: how to turn access into independence in a world where technology partnerships are increasingly instruments of power rather than pathways to parity.

That is it for the third essay. In the next essay, we will explore the question of GCC leverage in the AI race through its own competitive advantages. Make sure to subscribe. Thank you for reading. Make sure to subscribe to for more!

Essay 1:

Rocks, Rockets, & Robots

Hard Choices in Gulf AI Sovereignty?

Arab Gulf states face a strategic AI dilemma. The desire to cultivate AI sovereignty means hedging technology partnerships in an increasingly bifurcated global semiconductor competition between U.S. and Chinese ecosystems. The GCC finds itself positioned between two incompatible systems. Aligning with Washington’s semiconductor architecture offers acces…

Read more

4 months ago · Jesse Marks

Essay 2:

Rocks, Rockets, & Robots

The Decoupling Debate: Chinese Scholars Assess U.S. Chip Restrictions

I have listened at length to Washington-based debates over whether or not to provide chips to China. Washington asks: 'How much should we restrict?' Beijing asks: 'How quickly can we substitute?' This paper, the next in an essay series on debates around AI between the US, China, and Middle East, is an exploration into the other side of the chip debate i…

Read more

4 months ago · 1 like · Jesse Marks