Dario Amodei, co-founder and chief executive of Anthropic, said in an interview broadcast on Sunday that "the industry has lied" about the dangers of artificial intelligence, and called on governments and citizens to intervene.
Amodei told CBS News program Sunday Morning that real dangers exist, and that for a long time the industry has lied to people about the technology being risky. He said Anthropic had never done so, but that much of the rest of the industry had.
The Anthropic chief made the comments after calling on Saturday for a slowdown in the development of new AI models, to allow more time to understand the risks tied to the technology's capabilities. That call won public backing from OpenAI chief executive Sam Altman and from Elon Musk, founder of xAI.
Anthropic is the San Francisco based AI company behind the Claude chatbot, and is widely known for positioning itself as more safety focused than rivals such as OpenAI and Google. Amodei previously worked at OpenAI before co-founding Anthropic in 2021.
Warnings on cyberattacks and bioterrorism
In the CBS interview, Amodei warned of risks from the misuse of AI for potential cyberattacks and bioterrorism, from the loss of control over AI systems, and from economic disruption. Despite those warnings, he rejected the idea of a "total ban" on the technology.
He said he did not think banning it completely was the right approach. He said he did not like the idea that the incredible benefits, including medical cures, would never be realized, and added that the technology would be developed in other countries regardless, including authoritarian ones.
He described the toughest dilemma in the debate as the scenario in which the United States slows its AI development while China continues to advance.
Call for joint government oversight
Amodei said he felt uncomfortable that a private company, rather than a democratically elected government, was in charge of such a powerful tool. He argued for a combination of democratic governments to establish joint governance and monitoring of AI companies.
He said that if AI is built the right way, the probability of something bad happening is very low, but if it is built the wrong way, the probability of something bad happening is very high.
Trump and Johnson back continued development
Despite Amodei's warnings, United States President Donald Trump and House of Representatives Speaker Mike Johnson said on Sunday that AI development should continue, in order to prevent China from seizing the technological lead.
