Canadian Prime Minister Mark Carney warned that U.S. government restrictions on Anthropic’s AI models, Fable 5 and Mythos 5, demonstrate the risks of global overreliance on a small cluster of American technology providers. Following a directive from the Trump administration to limit access by foreign nationals, Anthropic pulled the models offline. Carney, speaking in Ireland ahead of the G7 summit, stated that nations must prioritize diversifying their artificial intelligence infrastructure to avoid similar disruptions to essential digital services.
Why are U.S. regulators restricting AI models?
The U.S. government implemented these export controls to address cybersecurity concerns regarding the capabilities of advanced artificial intelligence. According to Anthropic, the Mythos model is capable of identifying and exploiting complex computer vulnerabilities, often surpassing the performance of human cybersecurity experts. By forcing the company to take these models offline, the administration aims to prevent foreign actors from leveraging this technology for malicious cyber activity.
Anthropic’s Mythos model was announced on April 7 with the company explicitly stating it would limit access to select customers due to its “strikingly capable” nature.
How does overreliance on American AI affect global trade?
Prime Minister Carney linked the current AI restrictions to Canada’s broader economic strategy of diversifying trade beyond the United States. Currently, more than 70% of Canadian exports are directed toward U.S. markets. Carney has set an official target for Canada to double its non-U.S. exports over the next decade. He suggested that the “chill in investment” caused by ongoing trade tensions makes the concentration of AI power in a single country a strategic liability for international partners.
What is the future of AI governance at the G7?
Artificial intelligence is slated to be a primary discussion point among G7 leaders in Evian-les-Bains, France. Prime Minister Carney noted that while he held preliminary talks with French President Emmanuel Macron, there is no expectation of a simple, unified resolution. The complexity of AI safety and trade means that negotiators, including Canada’s Dominic LeBlanc and U.S. Treasury Secretary Scott Bessent, will focus on technical discussions rather than immediate policy declarations. Carney emphasized that “it is never a good idea to have one option” when building critical national infrastructure.
Comparison: Market Access vs. Security Controls
| Model | Access Status | Primary Driver |
|---|---|---|
| Fable 5 | Limited/Offline | Compliance with U.S. directive |
| Mythos 5 | Strictly Limited | Cybersecurity vulnerabilities |
Frequently Asked Questions
Why did Anthropic take Fable and Mythos offline?
Anthropic took the models offline to comply with a U.S. government directive aimed at preventing foreign nationals from accessing advanced AI technology that could be used for cyber exploitation.
What is Canada’s stance on AI development?
Prime Minister Mark Carney advocates for the diversification of AI providers, arguing that relying solely on U.S. companies creates systemic risks for international trade and national security.
Who is negotiating the USMCA renewal at the G7?
Discussions regarding the trade agreement involve Canada’s Dominic LeBlanc and Janice Charette, as well as U.S. Trade Ambassador Jamieson Greer and Treasury Secretary Scott Bessent.
Stay informed on the intersection of global policy and emerging technology. Subscribe to our newsletter for weekly updates on international trade and AI regulation.