G7 debate trusted partners access to cutting-edge US AI
G7 leaders are discussing a plan to give selected “trusted partners” access to advanced AI models from US firms such as Anthropic, potentially opening a path around restrictions on use by non-Americans, three diplomatic sources say.
On Friday, Anthropic disabled all users’ access to its most advanced AI models, Fable 5 and Mythos 5.
The company made the move after US President Donald Trump ordered Anthropic to block foreign nationals from accessing its most advanced artificial intelligence models, citing national security concerns.
Representatives of a number of countries attending the annual summit of leaders of the seven wealthy nations are discussing with US representatives the idea of expanding access to “advanced artificial intelligence” models, one of the diplomatic sources said.
This was mainly with US Commerce Secretary Howard Lutnick on the sidelines of the G7 launch dinner at the French lakeside resort of Evian-les-Bains on Monday.
A second source, who declined to be named because talks are ongoing, said “reliable partners” could be countries or companies.
An agreement providing broader access to advanced models would allow G7 countries to use them to develop stronger cybersecurity defenses against rivals such as China.
Trump’s White House official said in a statement that the president’s team “has an open line of communication with our allies, and we are committed to addressing national security concerns with the Anthropic model.”
AI executives from Anthropic, OpenAI and Google, which have all developed highly advanced models, are expected to attend a working lunch on Wednesday to talk about technology issues including regulation, AI infrastructure and networks, Reuters previously reported.
A spokesperson for Anthropic did not immediately respond to requests for comment.
Cybersecurity experts believe Anthropic’s Mythos, a model designed to find flaws in computer code, could accelerate attacks on banks’ technology systems. The European Union wants access to Mythos to examine the results of the model.
Before Trump’s order, Anthropic had given Mythos access to certain organizations in “more than 15 countries” so they could use the product to scan computer systems for vulnerabilities, according to a statement from the company.
The organizations included organizations in the healthcare, communications, energy and water sectors, according to the statement.
News of the “trusted partners” scheme was first reported by the Financial Times.
Australia’s Associated Press is the beating heart of Australian news. AAP is Australia’s only independent national news channel and has been providing accurate, reliable and fast-paced news content to the media industry, government and corporate sector for 85 years. We inform Australia.
