# Why Did Trump Declare 'Whoever Wins AI, Wins'?

According to Bloomberg Technology, President Donald Trump declared on Sept. 13, 2026, that 'whoever wins AI, wins.' That statement is not rhetorical flourish—it is a policy signal. It elevates AI from a technology sector to the central axis of US economic and national security strategy, and it implicitly tells every agency, ally, and chipmaker that alignment with American AI leadership is non-negotiable.

The framing matters because it changes the incentive structure. If AI is the scoreboard, then export controls, talent visas, and data rules are not trade policy—they are war-fighting tools. Bloomberg Technology reported that both the US and China see AI as the new focal point of their economic rivalry, and that neither trusts the other's motives. That mutual distrust is the engine of the race.

# What Are the Five Flashpoints—and Who Is Winning Each?

Bloomberg Technology's Sept. 29 explainer lays out five flashpoints: advanced chips, AI talent, data access, technical standards, and military AI. Each is a distinct battlefield with different dynamics. On chips, the US leads in design but depends on Asian fabrication; on talent, China produces more AI graduates but the US retains the top researchers; on data, China's scale and state access give it an edge in volume, while the US leads in diversity and quality; on standards, the EU and US are pushing rules-based frameworks while China exports its own model; on military AI, both are racing without guardrails.

The evidence supports a split verdict: the US leads in frontier model quality and chip design, China leads in deployment scale and manufacturing integration. Neither has a knockout blow. According to Bloomberg Technology, the question is whether the two nations can put aside their disagreements and cooperate in building a safe and sustainable AI environment—and the answer, based on their behavior, is no.

# Can the US and China Cooperate on AI Safety?

Short answer: not in any meaningful way before 2028. Bloomberg Technology reported that 'for now, neither trusts the other's motives, and neither wants to slow down.' That is the core obstacle. AI safety cooperation requires shared threat perception and mutual verification, and neither condition exists. The US views China's AI development as a security threat; China views US export controls as an attempt to strangle its tech sector.

There is a precedent for limited cooperation—the US and Soviet Union negotiated arms control during the Cold War—but AI is dual-use, diffuse, and fast-moving. Verification is nearly impossible, and the commercial incentives to cheat are enormous. I expect no binding bilateral AI safety treaty before 2028. Instead, expect unilateral safety frameworks and competing international standards.

# Who Gains and Who Loses From the AI Cold War?

The clearest winners are US chip designers like Nvidia and AMD, which benefit from export controls that lock out Chinese competitors, and Chinese AI champions like Huawei and Baidu, which gain a protected domestic market. The clearest losers are neutral countries and open-source researchers. Third countries like India, Brazil, and EU members face pressure to choose a side, and open-source AI communities are caught between US export rules and Chinese data regulations.

Bloomberg Technology's framing suggests the race is zero-sum in the minds of policymakers. That is dangerous. Zero-sum thinking leads to overinvestment in military AI and underinvestment in safety research. The market dynamic that changes: capital flows to national champions, not to the best ideas. That is a recipe for duplication and waste, but it is also the likely path.

# What Should We Expect Next?

Expect escalation, not de-escalation. The US will tighten chip controls; China will accelerate its domestic stack. The EU will try to carve out a third way but will lean toward US standards. The most likely near-term development is a US-led coalition on AI standards that excludes China, followed by a Chinese counter-coalition. According to Bloomberg Technology, the question of cooperation remains open, but the evidence points to a hardening split.

The one wild card is a major AI safety incident—a military AI accident or a catastrophic model failure—that forces both sides to the table. Absent that, the race continues.

# Predictions

1. By Q2 2027, the US Commerce Department will extend export controls to AI model weights and inference chips, targeting Chinese cloud providers.
2. By 2028, China will announce a domestic AI chip that matches Nvidia's H100 in training performance, reducing reliance on US hardware.
3. No binding US-China AI safety treaty will be signed before 2030; instead, the EU will broker a plurilateral safety accord excluding both the US and China.

# Article Summary

• Trump's 'whoever wins AI, wins' declaration on Sept. 13, 2026, formalized AI as the core of US-China rivalry.
• Bloomberg Technology identifies five flashpoints—chips, talent, data, standards, military AI—where the two powers are colliding.
• The US leads in frontier models and chip design; China leads in deployment scale and data volume.
• Cooperation on AI safety is unlikely before 2028 because neither side trusts the other's motives or wants to slow down.
• The biggest losers are neutral countries and open-source researchers caught between incompatible regulatory regimes.