Canadian Prime Minister Mark Carney says US AI restrictions underscore risks of dependence

Canadian Prime Minister Mark Carney criticized the U.S. government’s recent restrictions on advanced AI models as a warning sign of overdependence on American technology providers. The Trump administration ordered Anthropic, a San Francisco-based AI company, to take its newest models—Fable 5 and Mythos 5—offline to comply with export controls, limiting their use by foreign nationals. Anthropic had released Fable 5 publicly but restricted access to Mythos 5 due to its advanced capabilities, including surpassing human cybersecurity experts in identifying vulnerabilities. Carney, speaking in Ireland ahead of the G7 summit in Evian-les-Bains, France, framed the restrictions as evidence of the dangers of relying on a single AI provider. He stressed that while no wrongdoing occurred, the situation underscores the need for global diversification in AI development. ‘It is never a good idea to have one option,’ he said, linking the issue to Canada’s push to double non-U.S. exports over the next decade amid trade tensions. The prime minister also discussed AI with French President Emmanuel Macron, acknowledging that no immediate breakthroughs are expected from the summit due to the complexity of the challenges. Meanwhile, Canada’s USMCA negotiations will proceed through trade officials, with no direct meeting scheduled between Carney and Trump. Carney’s visit to his ancestral village in Ireland’s County Mayo highlighted local pride, with residents praising his advocacy for Canadian interests. Anthropic’s Mythos 5, announced on April 7, was described as ‘strikingly capable,’ prompting the company to limit access to select customers. The U.S. export controls mark the most significant effort yet to restrict access to cutting-edge AI, raising concerns about geopolitical fragmentation in technology development. Carney’s remarks reflect broader anxieties over AI governance and the risks of a fragmented global AI landscape.