The US-China AI Race 2026 Is a War No One Can Win

Pal Warns the AI Race Has No Clear Winner

Retired Goldman Sachs hedge fund manager and co-founder of financial media platform Real Vision, Raoul Pal, framed the deepening U.S.-China artificial intelligence (AI) competition in stark terms recently, stating:

“The U.S.-China AI race is a race no one can win and no one can afford to lose. Every great power competition in history was for territory, resources, or weapons. This one is the first that is for none of them. It is a race for the substrate of intelligence itself.”

His comments arrive as the AI race between the two largest economies has reached a critical juncture, with both nations pursuing radically different strategies. While the U.S. retains a clear lead at the technological frontier, particularly in compute scale, model performance, and large language model (LLM) development, China has pivoted toward a model built on efficiency gains, open-source diffusion, and deep integration of AI into physical-world systems.

A May 2026 analysis argued that China is now winning dimensions of the race that Western analysts had underweighted, specifically domestic AI deployment at scale, manufacturing integration, and the ability to build competitive models with significantly less compute than U.S. frontier labs require.

Rather than competing for a single AGI breakthrough, China has fragmented its strategy across multiple simultaneous races, be it model efficiency, AI adoption, or AI-controlled industrial systems.

Why Crypto Ownership and Universal Equity Matter

For Pal, the competitive stakes extend beyond pure technology into economic architecture. Speaking at Consensus 2026 in Miami, he proposed a concept called ‘Universal Basic Equity’ which gives citizens ownership stakes in AI systems as a structural response to the labor displacement expected as AI automates knowledge work at scale.

The proposal seems to align with Pal’s longstanding view that crypto-native ownership models may be better positioned than governments to distribute the economic gains from AI in the long run.

The broader geopolitical backdrop also carries implications for crypto markets in all of this given U.S.-China tech tensions have previously influenced export control regimes, chip access, and the regulatory environment for digital assets operating across both markets. A Brookings Institution analysis noted the competition spans multiple dimensions simultaneously (compute, models, adoption, integration, and deployment), making any single-axis assessment of “who is winning” incomplete.

What Pal’s framing adds to that picture is a philosophical dimension, i.e., the stakes may be unlike anything a geopolitical competition has involved before, since previous rivalries over territory, energy, or weapons were ultimately contests over finite resources. Intelligence and the systems that generate it are not analogous in the same way. That distinction, if Pal is right, can make the outcome of this race structurally different from anything that preceded it.