Anthropic CEO weighs in on whether AI could kill humans
What's the story
Dario Amodei, the CEO of Anthropic, has warned that artificial intelligence (AI) poses serious risks to humanity. He made the comments during an exclusive interview with CNN's Anderson Cooper. The warning comes after Jacob Coxon, a former researcher at Anthropic and OpenAI, resigned and raised alarms about the dangers of advanced AI systems. Following Coxon's resignation, another Anthropic safety researcher, Joe Benton, also left the company, voicing similar concerns.
Risk assessment
Amodei agrees with Coxon's concerns
Responding to a question about whether AI could potentially kill humans and how likely such an outcome might be, Amodei said he agreed with Coxon's concerns more than he disagreed.
He explained that while AI systems are becoming more powerful, the industry needs more time to implement adequate safety measures.
"What stops rival countries from racing ahead if Western firms slow down? If we take the right path, then the chance of something going wrong is very low," Amodei stressed.
Twitter Post
Here's what Coxon wrote
The people building AI earnestly believe that it could kill us all by the end of the decade. This is not a marketing stunt. If anything, many executives and senior researchers will couch their phrasing in the press to sound sensible - but I hear the same people express fear…
— Jacob Coxon (@hilbertspaess) September 9, 2026
Decision impact
Industry choices are crucial, not just probabilities
Amodei stressed that the focus should be on the choices made by the industry, rather than just probabilities.
He said, "I think it's more illuminating to think in terms of how do you decompose those probabilities? What are the paths in which things go well and what are the paths in which things go poorly."
This highlights his belief that we have agency over our future with AI.
Safety measures
Framework for safer AI development
In his blog post, 'We Must Pace the Frontier,' Amodei proposed a three-part framework to make AI development safer.
The plan includes independent oversight of advanced AI models during development, coordinated safety standards across the industry, and international cooperation on regulation.
He emphasized that each frontier AI company should commit to giving ongoing access to a team of embedded third-party evaluators whose role is to verify adherence to safety practices.
Industry response
Tech leaders agree on exploring ways to pace AI development
OpenAI CEO Sam Altman has also backed Amodei's proposal. He said his company is considering ways to pace the development of advanced AI systems.
"I agree with Dario that we need to pace the frontier," Altman said in a post on X. "Committing to having independent evaluators with employee-like access is a great idea, and we will do the same."
Following Altman's support, Elon Musk and DeepMind founder Demis Hassabis also backed Amodei's call.