AI regulation at G7: Macron urges common guardrails as US limits frontier models

French President Emmanuel Macron called on wealthy democracies to align on rules for advanced artificial intelligence. Macron spoke at a high-level Group of Seven meeting in France. The talks included senior AI executives and focused on safety and oversight. The push came as governments weighed how to manage powerful "frontier\" AI systems.

Representative image

The AI debate was coloured by a US move affecting Anthropic’s latest models. The Trump administration issued a directive last week barring foreign nationals from using them. Anthropic removed its newest systems, Fable 5 and Mythos 5, on Friday to comply. The dispute also fed European concerns about US control over AI ecosystems.

G7 AI regulation talks include top AI executives

Macron’s comments followed a G7 working lunch on \"Ensuring a safe, rapid and effective deployment of artificial intelligence.\" Attendees included OpenAI CEO Sam Altman, Google DeepMind CEO Demis Hassabis and Anthropic CEO Dario Amodei. Metas chief AI officer, Alexandr Wang, also joined, along with leaders of smaller AI labs.

Macron said it was positive that US officials accepted frontier AI could be risky. Macron also criticised the directive as a nationalist response. Asked if France and other G7 states requested access to Anthropic’s models, Macron said Macron urged the US not to keep cutting-edge AI for itself. Macron warned US firms could lose value if access is cut abruptly.

Macron urges AI regulation cooperation among democracies

Macron said democratic nations aimed to block authoritarian regimes from obtaining advanced AI. \"So let us move forward together,\" Macron said. \"Our relevant agencies must first cooperate so that, in the areas of security and cybersecurity, we have a smooth government-to-government relationship.\" Macron also said France would raise funding for its AI sector if cooperation fails.

Altman made a related appeal during the lunch attended by G7 leaders and many AI chiefs. Altman said AI’s future should be guided by people and democratic institutions. \"We need an international forum for discussion that establishes globally accepted standards for testing, provides expert and impartial analysis of capabilities and risks, and serves as a venue for cooperation among nations,\" he said.

G7 AI regulation proposals raise governance questions

Aidan Gomez, CEO of Canada’s Cohere AI, said participants discussed joint approaches to AI governance. \"I think the consensus was we need something,\" Gomez told The Associated Press. Gomez said democracies should ensure the G7 builds not only the most capable AI, but also the second most capable AI.

The lunch included France’s Mistral, Germany’s Black Forest Labs, Italy’s Domyn, Japan’s Sakana AI and UK-based Synthesia. The G7 includes France, the United States, Canada, Germany, Italy, Japan and the United Kingdom. Brazil, India, Kenya and South Korea joined some sessions as guest nations.

With inputs from PTI