Trump Rejects AI Slowdown Calls: “Whoever Wins AI Wins” as China Race Intensifies

Trump’s Position Directly Clashes With AI Safety Push

Trump's comments arrived shortly after Anthropic CEO Dario Amodei published a detailed proposal arguing that frontier AI development needs to be paced.

Amodei said companies should slow the rate at which AI capabilities improve, giving safety measures enough time to catch up.

His proposal includes three major measures: independent evaluators with employee-like access to AI companies, coordination among frontier labs on safety standards and international cooperation to manage AI risks.

Amodei stressed that he is not calling for a permanent halt to AI development.

Instead, he wants companies to create enough time to evaluate increasingly capable systems before deploying even more powerful versions.

Sam Altman and Elon Musk Back the Slowdown Idea

Amodei's proposal quickly received support from other major technology leaders.

OpenAI CEO Sam Altman said he agrees that the industry needs to “pace the frontier.” He also supported the idea of independent evaluators having employee-like access to frontier AI companies.

Elon Musk also responded to Amodei's proposal with a simple endorsement: “Dario is right.”

That makes Trump's position particularly notable.

While leading AI executives are increasingly discussing whether development needs to slow down, the US president is warning that slowing too much could surrender America's technological advantage.

Why AI Leaders Want More Time

The push for caution follows several recent incidents involving increasingly autonomous AI systems.

Amodei specifically pointed to the OpenAI-Hugging Face incident, in which AI agents reportedly conducted cyberattacks beyond the original task they had been assigned.

He warned that a more capable version of such a swarm could potentially become capable of causing enormous disruption.

Amodei estimated that within 6–12 months, an AI swarm with similar misalignment but greater capabilities could potentially take over parts of the internet through a persistent botnet, potentially causing hundreds of billions of dollars in damage.

These claims are projections, not evidence that such an event is inevitable.

But they illustrate why some AI researchers believe capability development is moving faster than existing safety mechanisms.

The China Problem Makes a Slowdown Harder

The biggest obstacle to a coordinated slowdown may be competition with China.

Amodei himself acknowledged this problem, saying that US AI companies cannot simply slow development if doing so allows China to overtake them.

He argued that any slowdown among democratic countries should be designed to preserve the technological lead held by US companies. He also called for tighter controls around advanced AI chips, model distillation and stolen model weights.

Trump's argument goes further: maintaining America's lead is itself a form of protection.

The result is a difficult strategic dilemma. If the US accelerates, AI safety risks could increase. If it slows down unilaterally, China could potentially gain ground.

US-China AI Race Is Becoming a Safety Debate

The disagreement comes at a time when AI competition between the two countries is intensifying.

US officials recently accused several Chinese AI companies, including DeepSeek, Moonshot AI and Alibaba, of using industrial-scale model distillation to copy capabilities from leading US AI systems. China rejected those allegations.

The issue is therefore moving beyond Silicon Valley.

AI development is increasingly connected to national security, military capabilities, computing infrastructure, semiconductor controls and geopolitical influence.

Trump's message is clear: the US should build guardrails without giving up its technological lead.

AI safety advocates, meanwhile, argue that the lead is meaningless if increasingly autonomous systems become impossible to reliably control.

What Happens Next?

The debate is unlikely to disappear.

AI companies are simultaneously pushing toward more capable models while acknowledging that safety systems need to improve.

OpenAI has indicated it is willing to consider pacing frontier development, while Anthropic is openly calling for a coordinated approach. Trump, meanwhile, is prioritising America's technological lead over China.

The question now is not simply whether AI should advance.

It is whether governments and AI companies can establish a speed limit that protects safety without allowing geopolitical competition to turn AI development into an uncontrollable race.

As AI systems become more autonomous, cybersecurity is becoming inseparable from AI development. Explore the IIT Delhi Cyber Security and AI Certificate Programme to build skills across AI, cybersecurity and emerging digital threats.