Anthropic CEO calls to ‘slow the pace’ of AI development amid safety concerns

Anthropic CEO Dario Amodei said Saturday that artificial intelligence development must slow down to allow for the implementation of necessary risk prevention safeguards.

“We must slow the pace at which we improve the capabilities of AI models,” Amodei wrote in a blog post . “Progress will still seem fast, and we must make wise use of the time we gain.”

Amodei recommended that AI companies immediately adopt measures aimed at “pacing” the development of their models, warning of potentially dire consequences if this technology is left unchecked.

“To be clear, pacing does not mean halting model training or technical progress, but ensuring companies take adequate time to align and safeguard their models, and for third party evaluators to confirm this,” Amodei wrote. “Our pacing framework is an attempt to further strengthen our commitment to safety and encourage a race to the top.”

Amodei’s announcement of this slowdown proposal follows stark warnings this week from several current and former employees of AI industry leaders about the dangers of this technology.

The Anthropic CEO proposed a three-pronged approach on Saturday.

The first step outlined in his pitch is to give external evaluators “employee-like access” to oversee the implementation of safety procedures and report incidents.

“Anthropic is unilaterally committing to this step now,” Amodei wrote. “We intend this to be part of a broader push to redouble efforts on our safety and alignment work.”

Amodei then called on AI companies and democratic governments around the globe to introduce widespread safety procedures. He also called on democratic nations to place pressure on authoritarian governments, like China, to agree to prohibit the development of “obviously dangerous uses” of this technology.

The Anthropic chief ended his Saturday blog post by emphasizing his belief in AI technology’s potential to “enormously improve the quality of human life.”

“My desire to achieve these benefits is undimmed,” he wrote. “But the benefits will only be achieved if we build the technology in the right way, and — so long as we use the time we gain well — it is worth taking unusually deliberate care to get it right.”

Amodei noted that “progress will still be relatively fast” during a potential slowdown period, but he urged AI companies to “use this time to advance” technology safeguards.

“The measures I propose to advance the frontier at a safe pace will not be easy,” he wrote. “But I believe we owe it to humanity to try.”

Sarah Heck, Anthropic’s head of policy, commended Amodei’s proposal as an “important call to action” and expressed the company’s interest in working with Congress to implement the CEO’s suggested guardrails.

“The government has a critical role to play here, including blocking the sale of the most advanced chips to adversarial nations like China, enacting a national law requiring testing of frontier models, with the power to block the most advanced models that prove to be unsafe,” Heck wrote on social media .

Rep. Don Beyer (D-Va.), who co-chairs the House AI Caucus and has introduced several bills aimed at addressing the risks of AI, called on lawmakers this week to take action amid a new wave of concern about the dangers of this technology.

“Doing nothing on AI is unacceptable and dangerous,” he wrote in a Wednesday social media post . “At a minimum we should be holding hearings, getting answers from the companies involved about these incidents, and passing those bills that are ready ASAP.”