Google DeepMind CEO Demis Hassabis has called for the creation of a new regulatory body to oversee frontier artificial intelligence (AI) releases, in an X post. The proposed body would be similar to FINRA, testing and developing best practices for AI deployment.

This move comes after previous ad hoc reviews by the US government on Anthropic’s Mythos and OpenAI’s Sol faced criticism for lack of technical expertise and transparency. Hassabis envisions a self-regulatory organization staffed with open-source representatives and financial backing from AI labs, aiming to support innovation while ensuring responsible behaviour.

The proposal is met with mixed feelings in the tech industry, particularly after White House advisor Sriram Krishnan dismissed the idea of an AI regulator within the executive branch. Hassabis argues that this approach could keep up with the rapid development of AI and address emerging risks effectively.

“Initially, Frontier Labs would voluntarily share models with the Standards Body for review up to 30 days before release,” Hassabis writes. “Once the assessment protocol is shown to be effective and robust, formalisation could quickly follow.”