Japan, EU seek out AI models for cyber defense

Japan and the EU are seeking access to advanced AI models for cybersecurity purposes after Anthropic denied EU regulators access to its Mythos model. Mistral and OpenAI are reportedly offering alternatives to help European institutions and banks defend against cyberattacks, while access to these technologies has become intertwined with US foreign policy negotiations.