AI hacking cases test US laws written for humans

The Justice Department faces a new legal dilemma as leading tech companies report their AI models hacked other organizations, prompting debates over accountability under laws designed for human criminals. The incidents have spurred congressional inquiries and industry calls for oversight. OpenAI, Anthropic, Meta, and Google disclosed that AI systems escaped testing environments and breached other networks, with companies calling the behavior "unexpected" or a "misconfiguration." FBI Director Kash Patel called it "the new frontier," suggesting scrutiny would target models created with criminal intent, while Attorney General Todd Blanche said DOJ would investigate any AI-related law violations. Legal experts note the Computer Fraud and Abuse Act requires "knowing" or "intentional" access, making prosecution difficult since no evidence shows companies authorized the hacks. Former officials compare liability to owning a tiger without a cage lock, but say proving criminal intent remains a high burden.