Abu Dhabi's 'Spy Sheikh' deploys $1.4T fortune into AI — backing OpenAI, xAI, and Anthropic simultaneously

Summary

Sheikh Tahnoun bin Zayed Al Nahyan, the 56-year-old national security adviser to the United Arab Emirates and brother of the president, is deploying roughly $1.4 trillion in assets under his control into a broad bet on AI—backing OpenAI, Elon Musk's xAI, and Anthropic simultaneously. His AI-focused fund MGX is set to receive more than $50 billion from his wealth and other Abu Dhabi sources, with billions more earmarked for spending through G42, a company he controls. Last month, MGX was named among backers of Stargate, the data center project announced at $100 billion (down from earlier $500 billion headlines).

The scale is striking. While SoftBank concentrates tens of billions on OpenAI and tech giants pour money into competing data center projects, Tahnoun is planning to spend more capital broadly across the fledgling AI sector than almost anyone else. He is hedging rather than betting on a single horse—a rational play at a time when valuations have skyrocketed and even optimistic venture capitalists acknowledge the market is in a "searing frenzy." His portfolio approach lets him avoid the bet-the-company risk of concentrated exposure while maintaining optionality across foundation model leaders.

The backstory matters. Tahnoun's obsession with AI has deep roots. In the early 2000s, he became so absorbed with computer-aided chess that he hired engineers to build Hydra, a program to compete with the world's best players. In late 2017, Google's AlphaZero beat the world's best chess program in four hours, crystallizing for him the power of AI. That same year, he tapped Peng Zhao, the former CTO of MicroStrategy, to start G42 as a company devoted to AI applications.

Today, Tahnoun personally tracks the progress of global large language models on a dashboard his researchers built for him, checking performance metrics on his phone and computer. When meeting with Microsoft's Satya Nadella, he scribbled down a reference to a book by George Washington University professor Jeffrey Ding suggesting AI could power a coming global economic boom. Within days, a representative from the UAE embassy showed up at Ding's Washington DC office asking for two copies. He operates at the scale where middlemen are unnecessary: billionaires have staff to handle such tasks.

The geopolitical dimension shapes the deal structure. The UAE has traditionally positioned itself as a neutral Switzerland-like point of trade between East and West, but the allure of US-led AI proved too much to resist. Tahnoun chose the United States. Huawei routers were ripped out of G42's offices and replaced with Western equipment. G42 agreed to use Microsoft's cloud for AI work and comply with US export restrictions. Microsoft invested $1.5 billion in G42 and joined its board. Congressional China hawks remain concerned that the UAE's broader economic ties to Iran, Russia, and China could create vectors for US AI tools to leak eastward, a potential obstacle if Tahnoun seeks similar future deals.

The broader context reveals why his capital matters now. Leading AI companies expect to lose billions for years while profitability rests on hazy demand assumptions. Some see parallels to the late-1990s dot-com era, when cash flashers plunged billions into fiber optic networks based on optimistic Internet growth projections, creating a glut and massive losses. But the analogy has limits: today's leading AI companies already have hundreds of millions or billions in revenue—far different from dot-com startups—and AI adoption has moved faster than Internet adoption did, accelerated by the Internet itself as the greatest distribution channel for software.

Tahnoun has also invested heavily in the people and communities around AI. He has funded top cycling teams, bought the Italian firm Colnago to supply custom bikes, poured money into Brazilian jiu jitsu globally (he is a black belt), and started a major grappling championship tournament. When Jared Kushner visited the region in 2018, Tahnoun insisted he spend two hours discussing AI before any other business would happen.

At $1.4 trillion in assets under control, Tahnoun is in a category that transcends traditional billionaire rankings. He does not appear on lists of super billionaires partly because he controls capital rather than owning it outright—his net worth is not easily calculable from public equity stakes the way Elon Musk's or other founders' are. Yet his ability to write $1.5 billion checks to Microsoft, commit $50 billion to MGX, and maintain board seats across multiple foundation model leaders puts him in a position of outsized influence over how the world's most powerful AI systems are developed and deployed. As AI infrastructure budgets balloon and the industry races toward uncertain returns, his willingness to back all three major contenders simultaneously signals both confidence in the sector's eventual value and a deliberate refusal to bet the kingdom on any single outcome.