Anthropic CEO Outlines Plan to Pace AI Development

Anthropic CEO Dario Amodei has called for a more cautious approach to artificial intelligence development, proposing a strategy to "pace the frontier." This move comes amid increasing warnings from AI researchers regarding the potential dangers of the technology. Amodei identified two primary drivers for this shift: the recent OpenAI-Hugging Face hack and the accelerating speed at which AI is advancing, specifically its growing capacity to help build subsequent generations of AI. He emphasized that while progress will still appear fast, it is essential to slow the improvement of model capabilities to make wise use of the available time. The proposal follows the resignation of researcher Jacob Coxon from Anthropic, who expressed concerns that leading AI firms are "gambling with our lives" and that the technology could pose an existential threat by the end of the decade. Support for the initiative has come from other industry leaders. OpenAI CEO Sam Altman stated that pacing the frontier has been a primary topic of internal discussions, and SpaceX CEO Elon Musk expressed his agreement. As a first step, Amodei suggested the use of "embedded evaluators" from third-party organizations, such as METR. These evaluators would verify that companies are adhering to their safety commitments and ensure that any safety incidents are properly reported.