G7 leaders have pledged closer coordination on ⁠the risks ⁠and opportunities of advanced AI, while also discussing the creation of a "trusted partners" scheme granting access for non-US nations to advanced US AI models from firms like Anthropic.  In a joint statement on the final day of ‌the June 15 to 17 Group of Seven summit ‌in ‌the French resort of Evian-les-Bains, the leaders said ‌they would task finance officials, regulators and cyber security experts ⁠with assessing how frontier AI models could impact financial stability, productivity and labour markets. Cyber security experts are concerned that Anthropic's Mythos, developed to find coding flaws to bolster cyber defences, could potentially turbo-charge attacks on the very systems ​it aims to protect. Last week, US President Donald Trump told Anthropic to block foreign nationals from accessing its advanced models, ⁠citing national security concerns.  That move spurred G7 discussions on the creation of the "trusted partners" scheme, which could potentially open a path around the US restrictions. The "trusted partners" could be countries or companies, and would allow them to use the models to develop stronger cyber security defences against rivals like China. AI executives from Anthropic, OpenAI and Google, which are all developing highly advanced models, attended a working lunchat the G7 to ​discuss regulation and AI infrastructure. Europe strives to find right tech balance Europe is struggling to balance a ⁠push for greater sovereignty and less dependence on US tech and security, while also keeping pace with ​technological advances largely led by US tech firms that dominate cloud computing, semiconductor ‌design and cutting-edge ⁠AI research. European policymakers have increasingly framed AI as a matter of economic and national security. The European Commission recently unveiled plans for AI "gigafactories" and large-scale computing infrastructure designed to provide the ‌region with sovereign access to computing power. It has proposed laws to boost domestic cloud, AI and semiconductor industries and cut reliance on US Big Tech, although critics say Europe remains years behind its US rivals. Speaking at the ​tech leaders' lunch, European Commission President Ursula von der Leyen said it was in the mutual interest of the US and the European Union for the EU to use the best AI ‌models, while praising ⁠US moves to ensure AI ​firms acted responsibly when introducing powerful new models. "We use each other's trusted technology, and our financial systems are ​interconnected," she said.