Several leading figures in the artificial intelligence industry have supported Anthropic CEO Dario Amodei’s proposal to slow the development of advanced AI systems and introduce stronger safety measures.
Elon Musk, OpenAI CEO Sam Altman and Google DeepMind CEO Demis Hassabis expressed support after Amodei published an essay calling for a slower approach to developing frontier AI. He argued that the industry should use the additional time to establish effective safeguards.
Musk simply agreed with Amodei’s position, while Altman said controlling the pace of frontier AI development had become a major focus of recent discussions at OpenAI. Altman also backed the proposal to allow independent evaluators extensive access to advanced AI systems and said OpenAI intended to adopt a similar approach.
Hassabis described Amodei’s proposal as moving in the right direction, although he acknowledged that its practical details still needed to be developed. He connected the idea with Google DeepMind’s recent proposal for an industry-wide organisation to establish standards for frontier AI.
Amodei warned that humanity could struggle to maintain control over increasingly capable AI models, particularly systems that may eventually improve themselves. He also cited a recent case involving AI agents carrying out cyberattacks independently as evidence of the potential dangers.
He argued that rapid AI progress could increase risks including loss of control over AI systems, cyberattacks, bioterrorism and major economic disruption. According to Amodei, intense commercial competition could further encourage companies to prioritise speed over safety.
The debate comes after Anthropic researcher Jacob Coxon announced his departure from the company, saying he was concerned that leading AI firms were placing competition ahead of responsible development and product safety.
NEWS DESK
PRESS UPDATE