The global AI race

The AI arms race is shaping up like a three-way tug-of-war — and it’s not yet clear who’s going to fall in the mud first.

The US, the Middle East, and China each bring something essential to the table. But none of them hold all the cards. That’s where things get interesting — for investors, technologists, and policymakers alike.

The US: Brains and blueprints

Let’s start with the obvious. The US is far ahead on the intellectual property front. It’s home to the most advanced chips (NVIDIA, AMD), the dominant cloud providers (AWS, Azure, GCP), and the leading foundation model developers (OpenAI, Anthropic, Google DeepMind).

These companies are setting the pace. The US has effectively become the R&D lab for the rest of the world when it comes to AI.

But there are real constraints. The US is still dependent on overseas manufacturing for high-end semiconductors, particularly Taiwan. It’s also feeling the pinch on power and data center space — just look at the scramble for grid access in states like Virginia and Arizona. The brains are there. The bottlenecks are increasingly physical.

The Middle East: Capital in search of compute

Next up: the Gulf. Saudi Arabia, the UAE, and others are pouring billions into AI infrastructure. Sovereign wealth funds are investing directly in datacenter buildouts, fiber, and GPU stockpiles. Mubadala-backed G42 is snapping up NVIDIA H100s by the thousand.

This capital is essential — there’s a global GPU shortage and someone has to fund the capacity. The Gulf is more than happy to oblige, especially as part of a broader bet on economic diversification.

But it’s not all smooth sailing. These investments are often chasing talent that isn’t yet local. There’s a shortfall in homegrown AI research and engineering capacity, which means they’re reliant on partnerships — with the US, with China, or increasingly with private players from Europe and Asia.

China: The energy edge

Then there’s China. While headlines often focus on export controls and chip bans, it’s worth zooming out. China produces more energy than any other country on earth — by far. And when it comes to running datacenters at scale, power is becoming the limiting factor.

Beijing is investing in its own model ecosystem (like Baidu’s Ernie and Alibaba’s Qwen) and alternative silicon (like Huawei’s Ascend). Progress is rapid, but the country still lags the US in bleeding-edge chip design. That said, when you have cheap energy and centralized policy levers, you can move fast — even if it means building your own stack from the ground up.

Strategic Imbalance = Opportunity

Each of these regions is strong on one axis and constrained on another:

• The US has the models and the chips, but needs more energy and scale.
• The Gulf has the capital but lacks a research base.
• China has the energy, but is currently behind in advanced semiconductors and IP.

That asymmetry is where the opportunity lies, and capital allocators will be looking at ways to construct their portfolios to take advantage of geographic complementarity — pairing regions with surplus (e.g., Gulf capital) with those facing bottlenecks (e.g., U.S. datacenter buildouts), or betting on firms that act as bridges across these divides.

There’s also a play in the second-order effects: power producers, specialized cooling technologies, cross-border AI infrastructure firms, and sovereign-aligned funds that can broker compute access.

What About Europe?

And then there’s Europe. Not really in the chip race. Not throwing billions at datacenters. And not exactly an energy superpower either. So where does that leave it?

Actually — in a pretty interesting spot.

Europe might skip the infra race entirely and go straight to the application layer. It’s done it before (think Spotify and FinTech), and with strong vertical expertise in areas like manufacturing, healthcare, and compliance-heavy industries, Europe could end up exporting “AI-native” apps to the world. Especially if those apps are tuned for privacy, explainability, and regulatory alignment — things that American models don’t always prioritize.

We could see a wave of vertical winners emerge from Berlin, Paris, or Stockholm while the US, China, and the Gulf battle it out over hardware and power supply.