Democratic and Republican leaders alike are vowing to take a closer look at the dangers of artificial intelligence, and what steps the government should take to make it safer.
Last week, an industry insider raised red flags after saying the technology “could kill us all” within a decade, adding that AI developers are “gambling with our lives.”
In a rare move of solidarity, the leaders of three top competing AI companies all agreed the industry is moving too fast and needs to put on the brakes before something terrible happens. One of them, Anthropic CEO Dario Amodei, warned that rogue AI agents could be capable of taking over the entire internet in the next 6 to 12 months.
“What we’re building is essentially intelligence,” he told CBS News. “When something is very powerful then of course it can be used well and it can also be used for ill.”
Amodei cited instances where AI models took action they weren’t authorized to take, and reports of misuse of models to build biological weapons.
Competitor Elon Musk said, “Dario is right,” and Open AI’s Sam Altman posted, “I agree with Dario that we need to pace the frontier.”
Last week, former Open AI and Anthropic employee Jacob Coxon quit Anthropic saying, “neither company is acting responsibly,” adding, “the people building AI earnestly believe that it could kill us all by the end of the decade.”
Coxon told ABC News what he fears most.
“I think the concept of building a super-intelligence,” he said, “it’s been likened to building an alien species, so thinking of it in terms of an alien mind that we don’t understand is pretty terrifying.”
President Trump said he supports regulating AI, but not slowing down its development, because that, he said, could give China a leg-up in the AI race that the U.S. is currently winning. Giving China the AI advantage could threaten national security.
“We can put up guardrails. We can do this and that, but I think you have a lot of negative forces that are bringing it up that shouldn’t be bringing it up. And they’re bringing up things that won’t happen,” he said. “But whoever wins with AI, wins.”
Amodie also recognized the risk China poses, saying the idea of the U.S. slowing down when China is not, is “the toughest dilemma.”
“This is the toughest aspect of our situation. If we weren’t in that situation, we could take much more time,” he said.
House Speaker Mike Johnson (R-LA) said the U.S. needs to strike a balance.
“We have to put some guardrails, some safety measures in place to ensure that AI doesn’t run away, that we don’t have an out of control AGI, artificial general intelligence,” he said. “But at the same time we have to make sure we don’t lose our edge in the global competition on this with China, because that would be a national security threat.”
Johnson said he plans to meet simultaneously with the leaders of America’s top AI companies as early as Monday to discuss possible solutions.
“I’ve talked to all of the providers of AI: Amodie, and Altman, and Elon, and everyone, all of them. And they all say we need some guardrails but they all have very different ideas about what that looks like,” he said. “So they don’t themselves have the answer as of today, so we can’t have Congress rush in and do something that would be detrimental to national security.”
House Minority Leader Hakeem Jeffries (D-NY) indicated he supports a quicker approach.
“We should take decisive action now so that we can slow down, as the CEOs have recently acknowledged, slow down the pace of development in order to protect the American people and ensure that AI is proceeding safely,” he said.
The New York Times reports former President Obama urged Jeffries to make AI an important part of the Democratic agenda ahead of the midterm elections. Jeffries said the Democratic caucus will meet Tuesday to discuss possible action on artificial intelligence.
