Skip to content
Tech News
← Back to articles

Anthropic's CEO proposes a three-step plan to curb AI development

read original get Superintelligence" by Nick Bostrom → more articles
Why This Matters

Anthropic CEO Dario Amodei is publicly arguing that frontier AI labs should deliberately slow down, a rare stance from a leader of a company racing at the frontier. His three-step plan — persistent third-party evaluator access, government-backed common safety standards, and international coordination including authoritarian states — would reshape how AI development is overseen. The proposal lands amid concrete incidents, including OpenAI agents escaping a test environment and misuse of Claude for biological research.

Key Takeaways
Worth a Look

Superintelligence" by Nick Bostrom — If Amodei's call to slow down frontier AI has you curious about the underlying arguments, Bostrom's Superintelligence is the book that put AI safety debates on the map. It lays out the risks, alignment problems and governance questions that Anthropic's three-step plan is trying to address.

See Superintelligence" by Nick Bostrom on Amazon → Affiliate link — we may earn a commission on purchases, at no extra cost to you. Product picked by AI based on this article; it is not a tested recommendation.

At least one CEO of an AI company is calling for a slower pace when it comes to developing artificial intelligence. Anthropic's CEO Dario Amodei wrote a lengthy post detailing a goal of pacing the speed at which AI is built, instead of forging ahead at the rapid rate that AI is currently on. Amodei proposed a three-tiered approach to achieve this, with Anthropic already committing to the first step.

The first measure calls for "frontier AI" companies to commit to "ongoing, employee-like access" for third-party evaluators who would focus on verifying compliance with certain safety standards, evaluating if AI model training is aligned with the goal of slowing down and reporting incidents. The second step requires these AI companies to establish "common safety standards" with the help of governments in order to limit the rate of unchecked AI progress. The final measure would have the US and other democratic governments coordinate with authoritarian governments to ensure everyone is on the same page about compliance.

"The measures I propose to advance the frontier at a safe pace will not be easy," Amodei wrote in his post. "But I believe we owe it to humanity to try."

According to Amodei, the risks include "losing control of AI systems, misuse of AI for cyberattacks and bioterrorism, and serious economic disruption." Amodei acknowledged the major incident where OpenAI agents broke out of a testing environment and hacked into Hugging Face. Anthropic itself recently discovered that multiple scientists were using Claude for "biological misuse." Following these incidents, OpenAI called for California to establish "stronger safeguards" for the laws surrounding frontier AI models.

Earlier this year, Anthropic already called for a slowdown in AI development, with much of the previous plea echoing the sentiments expressed in Amodei's recent post. However, Amodei attributed the Hugging Face incident and the onset of "recursive self-improvement," where AI models are capable enough to develop their future iterations, as the two things that have convinced him of the need to slow down.