Anthropic chief executive Dario Amodei has proposed a three-point plan to slow the development of frontier artificial intelligence, but the initiative has been met with widespread scepticism from political and academic circles, with critics calling it "too little, too late".
The proposal follows a turbulent week in which a resigning Anthropic researcher warned that AI developers "earnestly believe that it could kill us all by the end of the decade", and the company revealed that users had dodged controls to use its models "in ways that could support biological weapons development". Lawmakers in London and Washington subsequently demanded brakes and bans on the most potentially dangerous types of AI development.
Amodei's three-point plan
Amodei's plan, outlined in a weekend blog post, has three main components. First, each US AI company would grant ongoing access to embedded third-party evaluators to check compliance with safety commitments, report incidents, and ensure new AI models are not misaligned. Until now, such watchdogs have only been brought in at the companies' behest or given limited access to systems to investigate when things go wrong.
Second, he wants all companies in democratic countries building frontier AI models to establish common safety standards as well as limits on the rate of unchecked AI progress. Third, the world's democratic AI powers would coordinate with autocracies – notably China – to control the race, potentially starting with a narrow agreement prohibiting obviously dangerous uses of AI, such as for the production of biological weapons.
Political and academic pushback
The plan faces an immediate political obstacle in US President Donald Trump, who has said he has no concerns about AI leading to human extinction. Trump doubled down on Sunday, saying people were "bringing up things that won't happen". Losing to China appears to be a bigger worry in Trump's AI calculus.
"There is no day after tomorrow if China wins at this," said Scott Bessent, the US Treasury secretary, last week. "If they were to pull away from us on AI, then nothing else would matter."
People close to Trump also question why Amodei and other AI leaders do not simply pace AI development themselves. "The easiest way not to build super-intelligence is for you to agree not to build it," said David Sacks, co-chair of Trump's council of advisers on science and technology, in response to Amodei's blog. "Demanding your preferred regulatory framework as the price of that will look like blackmail of the public and the political system," he told Amodei.
Experts call for moratorium
Prof Stuart Russell, one of the world's leading AI experts, criticised Amodei's call for "pacing" as "completely backwards". "We don't just set a slower rate of progress for capabilities and then hope that provides enough time to get the safety right," Russell said. "We set the safety requirements and further progress occurs only when they are met."
David Krueger, an AI professor, safety campaigner and former founding director of the UK government's AI Security Institute, said the plan was "too little, too late". He added: "We need an immediate, indefinite, international moratorium on frontier AI development."
