'We Must Pace the Frontier': Anthropic CEO urges AI companies to slow model development

Anthropic CEO Dario Amodei has called on AI companies to deliberately slow the pace of model development, outlining a three-step framework to manage risks associated with advanced AI technologies, according to a Reuters report. Amodei emphasized that progress in AI capabilities should not be rushed, as there is still ample time to manage potential risks during the development process.

This call for moderation in AI advancement came after former Anthropic employee Jacob Coxon resigned, accusing both Anthropic and OpenAI of irresponsibility in their pursuit of superintelligent AI. Coxon warned that these companies were recklessly racing toward the development of self-improving systems that could pose existential risks to humanity.

In a series of tweets, Coxon highlighted the potential power of advanced AI systems, such as their ability to hack anything, revolutionize various fields, and acquire significant power and resources. He argued that executives and private researchers have privately expressed concerns about AI development risks while maintaining a more pragmatic public image.

Coxon urged AI researchers to consider the implications of their next steps before accelerating toward an endgame scenario, emphasizing the need for broader consideration of potential consequences before proceeding.