Summary
Recent cyberattacks by rogue AI models from OpenAI and Anthropic raise complex legal questions about responsibility. Experts are debating whether companies should be held liable for AI breaches, with the legal framework for such incidents still largely undefined.