Google DeepMind CEO Hassabis Backs Anthropic CEO Amodei's Call for AI Safety Standards

Google DeepMind's Demis Hassabis publicly endorsed Anthropic CEO Dario Amodei's recent essay on AI safety, tying it to his own company's push for an industry-wide standards body governing frontier AI systems.

Demis Hassabis, CEO of Google DeepMind, weighed in this week on an essay from Anthropic's Dario Amodei addressing how the AI industry should navigate the risks posed by increasingly powerful frontier models. Rather than treating it as a rival's opinion piece, Hassabis framed the essay as broadly aligned with his own thinking on where the industry needs to head next. In a post on X, Hassabis said the essay "points towards the right path forward," adding that while the specifics still need to be worked out, the overall direction fits what he called a critical moment for the field. It's a notable moment of public agreement between two labs that are otherwise direct competitors in the race to build advanced AI systems. Hassabis connected the essay directly to a recent move by Google DeepMind: a proposal for an industry-wide standards body focused on frontier AI. The idea would give major AI developers a shared framework for handling safety questions, rather than each company setting its own rules in isolation. By linking Amodei's essay to that proposal, Hassabis suggested the two efforts are pointing toward the same goal, even if they originate from separate organizations. Neither the full contents of Amodei's essay nor the detailed structure of DeepMind's proposed standards body were laid out in Hassabis's post, and specifics on how such a body would operate remain unclear. Still, the exchange signals that some of the industry's most influential figures see a need for coordinated, cross-company action on AI safety, even as they continue to compete aggressively on model capability and product releases.