After US blocks Mythos and Fable, Anthropic chief urges world leaders to not splinter over AI

Anthropic CEO Dario Amodei urged world leaders to not splinter over building advanced AI tools during the G7 summit in France. Amodei's comments come just days after the US banned access to Anthropic's Mythos 5 and Fable 5 models for all foreigners.

In Short

• Dario Amodei tells world leaders to resist temptation to splinter over AI access
• Amodei calls for world leaders to work together in AI race
• His comments come after US banned Mythos and Fable models for all foreigners

During the final day of the G7 summit in France, world leaders discussed the future of AI with some of the biggest names in the industry, including Anthropic CEO Dario Amodei, and OpenAI chief Sam Altman. Dario reportedly urged world leaders to not splinter over advanced AI tools just days after the US banned his company’s Mythos 5 and Fable 5 models for all non-US citizens.

As per a report from the Financial Times, the Anthropic chief told G7 leaders to “resist the temptation to splinter” over AI tools during a working lunch in vian-les-Bains, France, on Wednesday. He reportedly made these comments in front of US President Donald Trump.

That is, Dario Amodei wants democratic countries to stick together when it comes to powerful AI tools which can do a lot of damage if they fall into the wrong hands.

Dario Amodei’s remarks come at a time when the US’ move to block access to Mythos and Fable has sparked debate over the future of AI geopolitically. Some believe that countries must reduce dependence on foreign AI tools as a country like the US can easily block access by flipping a switch.

Sam Altman, Demis Hassabis support Dario Amodei’s comments

OpenAI CEO Sam Altman and Dario Amodei are not known to have the best of relations. During the India AI Impact Summit in February this year, the two infamously refused to hold hands together on stage. However, when it comes to the topic of countries working together in the AI space, it seems that Dario and Sam have similar views.

As per the report, Sam Altman agreed with Dario Amodei, and claimed that all countries that were attending the summit, including India, should have access to that cyberdefence tools.

Google DeepMind chief Demis Hassabis also joined the discussion. The three AI leaders insisted that the US must lead a collaborative effort on the development of AI models, while also pointing out potential risks in terms of bioterrorism and cyber security if democratic alliances fractured in the AI race.

What are world leaders saying?

The report added that French President Emmanuel Macron was quick to point out that if the US “from one day to the next can turn off the switch”, it would hurt the multitrillion-dollar American companies leading the AI race.

Macron also called for “stronger regulation of AI” and warned against “non-cooperation among democracies.” He added that what would be built in the coming months would be “a platform for discussion and co-operation among a number of democracies” to define common standards together.

Prime Minister Narendra Modi also raised concern over the US decision, as per the report. He reportedly stated that democratic nations must have access to AI models to protect critical infrastructure.

Keep in mind that Anthropic says that it is holding talks with the US government to revert this ban on access to Mythos and Fable. The two models were banned for all foreigners, including Anthropic’s own employees, over fears that they may be exploited to use for cyberattacks after Amazon researchers reported that the models could be jailbroken with certain prompts.

Reports also indicate that the US feared China-linked groups may get access to Mythos and Fable, which could then potentially use them for cyberattacks. Though Anthropic states that both models come with safeguards that prevent any potential misuse.

- Ends