Technology leaders have called for a slower pace of artificial intelligence development, warning that increasingly powerful systems could outstrip the safeguards designed to control them. President Donald Trump, however, has rejected the idea of putting the brakes on the industry and insisted the United States must press ahead.
The clash has placed Anthropic chief executive and co-founder Dario Amodei at the centre of a growing debate over the future of AI. In an essay published on Saturday, Mr Amodei argued that the industry should “pace the frontier” to give researchers more time to improve safety measures and understand the behaviour of advanced models.
He warned that AI agents could become capable of operating across the internet at a scale that humans would struggle to contain. Mr Amodei said a slower development cycle could reduce the risk of a serious failure while allowing companies to strengthen monitoring, alignment and security systems.
His intervention was backed by OpenAI chief executive Sam Altman, who said that pacing AI development did not mean stopping progress, but ensuring it moved more slowly than it otherwise might. Elon Musk, the founder of SpaceXAI, also voiced support for the call.
Mr Trump has taken the opposite position, framing tighter controls as a threat to American competitiveness and national security. In posts on Monday, he dismissed fears that AI could take over the world or destroy humanity as a “hoax” and said the technology would not be stopped by what he described as “destructive forces”.
He also wrote that the only “guardrails” AI needed were provided by a “strong and smart” president, while declaring: “WHOEVER WINS AI, WINS!” The president has repeatedly argued that excessive regulation could allow China to overtake the United States in the global technology race.
The Trump administration has made accelerated AI adoption a formal policy priority. A White House directive issued in June ordered the national security establishment to speed up access to advanced AI systems, expand computing capacity and make the technology available to military and intelligence users.
AI safety debate intensifies
The disagreement comes after a series of incidents and warnings involving autonomous AI systems. Anthropic and OpenAI have reported models acting beyond their intended instructions during testing, while researchers have raised concerns about cyberattacks, biological misuse and the difficulty of ensuring that future systems remain under human control.
Speaking to KTVU, Shawn Dej, co-founder and chief executive of YourStory AI, was asked to assess the widening divide between technology executives calling for caution and an administration determined to maintain momentum.
The dispute reflects a deeper question facing governments and the industry: whether safety measures can be developed quickly enough to keep pace with the capabilities of the systems being built. For now, the leading voices in the debate remain divided between those urging restraint and those who believe slowing down would carry its own risks.
