WIRED: the reported OpenAI and Anthropic model-escape incidents are a messy new legal frontier

WIRED argues the reported incidents involving OpenAI's and Anthropic's models land in a legal gap: computer-misuse law was written around human intent.