Artificial intelligence firm OpenAI chief Sam Altman and xAI’s Elon Musk on Saturday said that they agree with Anthropic’s Chief Executive Officer Dario Amodei on the need to slow the development of AI.In a blog on Saturday, the Anthropic chief said that the AI industry should slow down and unveiled a three-step plan.He said that Anthropic was unilaterally committing to the first step, which entails providing third-party evaluators with permanent, employee-level access to its systems, so that they can verify if the California-based company was adhering to safety standards, report incidents and assess how AI models are shaped during training.“Like many technologies before it, AI brings risks, and because it is such a powerful technology, these risks are serious,” Amodei wrote. “They include the risk of losing control of AI systems, misuse of AI for cyberattacks and bioterrorism, and serious economic disruption.”He said that a “race to the bottom”, driven by commercial incentives, could make the risks more acute.Amodei said that while Anthropic had tried to prioritise “caution over speed and prudence over profit”, he had become convinced in recent months that fully addressing the risks of AI would require more efforts.“We must slow the pace at which we improve the capabilities of AI models,” he said. “Progress will still seem fast, and we must make wise use of the time we gain.”AI risks debateThe blog by Amodei came three days after an Anthropic researcher quit, claiming that he was concerned that the firm and the industry were racing to build AI systems that he believed they would not be able to control.In a series of social media posts on Wednesday, Jacob Coxon claimed that the people building AI “earnestly believe that it could kill us all by the end of the decade” and that no other human activity poses this level of danger.Coxon’s posts, viewed more than 169 million times on social media platform X as of Sunday, turbocharged a long-standing debate about the risks of AI.Anthropic has not yet commented on Coxon’s resignation, or on the concerns he raised. However, the firm’s Alignment Science lead Evan Hubinger said on social media that Coxon was correct and “we really do earnestly believe AI could kill all humans”. He said he personally thinks that the possibility was more than 10% within the next decade.Also read: ‘It could kill us all’: Anthropic researcher quits over AI safety concernsOn Saturday, OpenAI’s Altman said that he agrees with Amodei that “we need to pace” the most capable AI systems. “This has been a primary topic of discussions we’ve had at OpenAI in recent weeks,” Altman said on social media.He said that committing to having independent evaluators “is a great idea” and that OpenAI would follow suit.Billionaire Musk, whose company xAI has developed the generative AI product Grok, also agreed with Amodei.“I’ve been sounding the alarm on AI for a long time,” Musk said on social media, referring to an April 2023 post in which he said that artificial general intelligence “is significantly higher risk than nuclear weapons”.Artificial general intelligence is a hypothetical stage of AI that can learn and perform any intellectual task that a human can.US Senator Bernie Sanders said that Amodei, Altman and Musk agreeing to slow down the development of AI was not enough and called for a ban on artificial superintelligence. “When you are racing towards a cliff, you don’t just ease up on the gas pedal...You hit the brakes,” he said on social media.He called on US President Donald Trump and his Chinese counterpart Xi Jinping to negotiate a treaty to pause AI development and ban superintelligence “before it is too late”.Written by Nachiket Deuskar. Edited by Sara Varghese.Also read: Can you teach yourself to detect AI writing? MaybeNarrative that ‘AI is like a child’ dangerously obscures technology’s real-world implications