source: techcrunch ai: deepmind ceo calls for an independent standards body to regulate frontier ai
level: business
google deepmind ceo demis hassabis called for a new regulatory body to oversee frontier ai model releases. in a post on x, he proposed a standards body modeled after the financial industry regulatory authority, or finra. this group would test models and develop best practices. initially, ai labs would voluntarily share models for review up to 30 days before release. once the assessment proves effective, it could become mandatory for deployment in the us market. labs would also work with the body to fix post-release vulnerabilities.
the idea builds on past ad hoc reviews by the us government of models from anthropic and openai. those reviews faced criticism for lacking technical expertise and clear decision-making. hassabis's plan would hand off those decisions to an independent organization. it would be backed by the government but funded by the ai industry. this self-regulatory approach could address concerns about government overreach. white house ai advisor sriram krishnan recently said there will not be an fda for ai, making a finra-like model more politically viable.
the proposed body would include open-source representatives and technical experts from the industry. it could outsource some evaluations to specialized ai safety groups. hassabis argues this approach is technically focused, supports innovation, and encourages responsible behavior. it is designed to adapt to new risks as they emerge and could be strengthened if needed. the goal is to keep pace with rapid ai advances while ensuring safety.
why it matters: a standardized review process could help manage risks from powerful ai models without stifling innovation, offering a practical path for industry-led regulation.
source: techcrunch ai: deepmind ceo calls for an independent standards body to regulate frontier ai