Meta’s updated commercial terms place broad responsibility for its AI features and resulting content on business users, with continued use after November 20, 2026 meaning agreement to take sole “[responsibility] for and assume any and all risks associated with” Meta’s AI features. The article says this defense gets harder to laugh off as AI agents gain more autonomy, because when an agent can make decisions, invoke tools, generate code, or interact with other systems, pointing at the technology after something goes wrong does not answer who had control and who gets stuck with the consequences. A new bipartisan proposal for an AI Agent Accountability Act would extend liability under the Computer Fraud and Abuse Act to AI agent operators and developers that recklessly cause damage and, in some circumstances, developers that fail to implement reasonable safeguards; the bill is only a proposal. The article’s new report, Contracts Are The AI Guardrails With Teeth, says AI governance policies and identifying AI risk are not enough to ensure accountability when third-party AI goes wrong, and that contracts are critical to connect governance intent with risk management execution. Before agreeing to terms on third-party AI, the article advises bounding the agent’s autonomy, demanding evidence such as logs, material change notifications, incident reporting, and investigation support, and matching accountability to control rather than accepting blanket risk transfers when the supplier controls the model, its safeguards, or changes customers cannot see.