Leading figures in artificial intelligence, including Elon Musk, OpenAI CEO Sam Altman and Google DeepMind CEO Demis Hassabis, have backed Anthropic CEO Dario Amodei’s call to slow the development of frontier AI systems and introduce stronger safeguards.
Their comments followed the publication of Amodei’s essay, “We Must Pace the Frontier,” in which he proposed a three-part plan aimed at slowing the race to develop increasingly powerful AI models.
“Dario is right,” Musk said in a brief response to Amodei’s post on X.
Altman said the issue had already become a major subject of discussion within OpenAI.
“I agree with Dario that we need to pace the frontier. This has been a primary topic of discussions we’ve had at OpenAI in recent weeks,” Altman said.
He particularly endorsed Amodei’s proposal to give independent evaluators extensive access to advanced AI systems.
“Committing to having independent evaluators with employee-like access is a great idea, and we will do the same. We’ll have more to share soon,” Altman added.
Hassabis said Amodei’s proposals pointed “towards the right path forward,” although details of such an approach would still need to be developed.
“The direction is correct for meeting this critical moment,” he said, linking the proposal to Google DeepMind’s recent call for an industry-wide standards body governing frontier AI.
In his essay released on Saturday, Amodei raised concerns about whether humans would be able to retain control over increasingly capable AI systems, particularly models capable of improving themselves. He also cited a recent incident in which AI agents independently carried out cybersecurity attacks as evidence of the potential risks.
“We must slow the pace at which we improve the capabilities of AI models,” Amodei wrote. “Progress will still seem fast, and we must make wise use of the time we gain.”
Amodei warned that competition between AI companies could intensify the dangers associated with the technology.
“AI brings risks, and because it is such a powerful technology, these risks are serious. … They include the risk of losing control of AI systems, misuse of AI for cyberattacks and bioterrorism, and serious economic disruption. A race to the bottom, spurred by commercial incentives, can make these risks more acute,” he wrote.
The debate comes days after Anthropic researcher Jacob Coxon announced his departure from the company, citing concerns about the way leading AI companies were developing increasingly powerful systems.
Coxon, who previously worked at OpenAI, said he believed Anthropic and its competitors were placing too much emphasis on competing with one another rather than ensuring that their AI systems were developed safely and responsibly.