Demis Hassabis Calls for AI Watchdog to Regulate Frontier Models

Demis Hassabis, the CEO and cofounder of Google DeepMind, has proposed the establishment of an AI watchdog with the authority to intervene if cutting-edge AI models pose a risk.

In a recent blog post, Hassabis suggested that the United States should take the lead in creating this oversight body, citing the country’s economic and technological influence as key factors. The proposed organization, similar to existing regulatory bodies like the Financial Industry Regulatory Authority, would consist of independent experts and representatives from open-source communities. Its primary role would be to assess new AI models before their release and implement industry-wide restrictions if deemed necessary.

Hassabis emphasized the growing importance of global regulation as AI technology advances, predicting that Artificial General Intelligence (AGI) could become a reality in the near future. He referred to this period as the “foothills of the singularity,” marking a significant shift in humanity’s technological evolution.

According to reports, Hassabis has been garnering support for his proposal from various stakeholders, including the Trump administration, AI research labs, and European officials. He aims to establish the AI oversight body before the end of the year and has received positive feedback from the current administration.

This initiative is part of a broader effort by Hassabis and other industry leaders to develop a comprehensive framework for governing AI technologies and mitigating potential risks. Despite the lack of global or national regulations specific to AI, Hassabis has been advocating for stricter controls, including measures to prevent AI-related bioweapons production.

Hassabis’ advocacy for AI governance aligns with recent calls from prominent economists and tech figures, such as Jack Clark from Anthropic and former Google CEO Eric Schmidt, urging world leaders to address the economic implications of AI advancements.