- Michael Kratsios advocates for lighter AI regulation via the Carolina Principles
- Under the system, existing laws would be applied before new laws are created
- China has reportedly signed up despite the scheme aligning with US interests
Speaking at a recent G20 meeting, US officials have urged other governments to rely on existing legal frameworks wherever possible rather than treating AI as a totally new policy problem in a bid not to lose competitiveness.
US technology advisor Michael Kratsios said countries signing up to the proposed Carolina Principles would be in agreement not to write entirely new regulatory systems specifically for AI.
The Principles would only support new rules where AI creates genuinely "novel" situations where existing laws can't be applied.
Carolina Principles push for lighter AI regulation
"Policymakers do not need to approach each innovation in isolation and should not treat every emerging technology as a first-of-its-kind policy problem," Kratsios said.
According to Reuters, China has already signed up to the Principles. While the country has not yet confirmed whether this is true, Kratsios did describe a meeting with Chinese science and technology minister Yin Hejun as "great."
Speaking at the event, SpaceX CEO Elon Musk called out current European regulation for harming technological progress. He also touched on the power needs of AI data centers as the sector continues to grow.
Despite the general consensus that lighter regulation could prevent setbacks, Google DeepMind CEO Demis Hassabis urged G20 governments to establish safety testing for AI systems, implying that at least some central oversight could be beneficial for the greater good.
This regulatory debate, which continues to play out, is also closely connected to ongoing US-China tensions, despite the apparent collaboration with China. It forms part of a geopolitical strategy that could keep US companies competitive. The position also closely matches the interests of major American AI companies, which would likely favor lighter regulation and fewer restrictions.
Reuters also reports that OpenAI CEO Sam Altman, Nvidia CEO Jensen Huang and Anthropic co-founder Tom Brown are also set to address delegates as G20 meetings continue.
Follow TechRadar on Google News and add us as a preferred source to get our expert news, reviews, and opinion in your feeds.
