Anthropic CEO Dario Amodei said Sunday that the United States should consider opening arms-control-style negotiations with China over artificial intelligence, drawing a direct comparison to Cold War-era talks between the U.S. and the Soviet Union.
Speaking to CBS Sunday Morning correspondent Jo Ling Kent, Amodei said both countries share an interest in avoiding the kind of runaway risk AI could pose, even as competition between the two nations intensifies.
“It’s not in any government’s interest to have bioterrorism,” Amodei said. “And so we should talk with the Chinese and see if we can work together to put limits to put a speed limit on the rate of this technology.”
Amodei, who describes himself as generally hawkish toward China, said the comparison to the nuclear standoff with the Soviet Union still applies despite deep political differences between the countries.
“For all our differences, we recognized that that was too many, that we needed to engage in talks that could lower the scale of the potential destruction,” Amodei said, referring to the tens of thousands of nuclear missiles once held by each side.
When Kent asked whether the Cold War should serve as a direct model for handling the AI race, Amodei clarified his position.
“I didn’t say we should use the Cold War as a model… disarmament negotiations like arms limitation, I do think that is an appropriate lens through which to think about this at the geopolitical level,” he said. “If we look at what AI models can do in terms of cyber, like they absolutely have geopolitical and national security significance. Of course, they do.”
The interview followed an essay Amodei published Saturday warning that AI capabilities are advancing faster than humanity’s ability to control them. That essay drew support from prominent tech figures including SpaceX and xAI’s Elon Musk and OpenAI’s Sam Altman, but it also drew sharp criticism from others in the industry.
Former White House AI adviser David Sacks dismissed warnings like Amodei’s as a tactic to justify heavier government control of the technology.
“[The warnings are] designed to scare the public about AI so that these NGOs can implement their preferred approach to AI, which is a government takeover,” Sacks said on “Special Report” last week.
Former OpenAI researcher Daniel Kokotajlo took the opposite view, telling Fox News Live on Saturday that voluntary safety measures by AI companies may not be sufficient to prevent advanced systems from slipping out of human control.
“After they’ve accumulated all of the hard power in the world, they run the factories, they’re integrated into the military… and humans are just sort of giving them orders,” Kokotajlo said. “Then they stop taking orders at some points, and they kill off the humans.”
Fox News Digital said it had reached out to Anthropic for further comment.
You must be logged in to post a comment Login