FTC Chair Warns AI Agent Deployments Face Liability for Harm and Nondisclosure

What happened

Why it matters

• ·Companies that deploy AI agents without clear disclosure face immediate exposure under existing FTC deceptive practice authority, with no new rulemaking needed to trigger enforcement. The FTC has used this authority against AI-related conduct before, as seen in the FTC Artificial Intelligence Compliance Plan.
• ·The chair's rejection of the 'independent actor' framing means enterprises cannot attribute agent misbehavior to the AI system itself. Responsibility flows to the deploying company, raising the stakes for governance gaps in agent oversight and action logging.
• ·Organizations in consumer-facing sectors, including financial services, retail, healthcare, and telecom, face the most immediate exposure. Any agent that contacts consumers, makes decisions affecting them, or takes actions on their behalf without disclosure is a potential enforcement target.

Governance controls affected

What to do now

• ☐Inventory every AI agent that contacts consumers or takes actions on their behalf, and confirm each one has a disclosure mechanism that tells consumers they are interacting with or being affected by an AI system.
• ☐Review customer-facing scripts, interfaces, and automated workflows to identify any point where an agent acts without the consumer knowing it is AI-driven, and assess whether that gap would satisfy an FTC 'deceptive practice' standard.
• ☐Map which employees or teams are accountable for each deployed agent's behavior, and document that accountability in writing so the organization can demonstrate human responsibility rather than pointing to the AI system.
• ☐Pull logs for the past 90 days of agent actions in consumer-facing workflows and confirm those logs are complete enough to reconstruct what the agent did, when, and on whose behalf in the event of an FTC inquiry.
• ☐Brief legal counsel and the compliance committee on the FTC chair's enforcement signal and assess whether any current agent deployment would warrant a voluntary disclosure review before regulators make contact.

What to watch next

Compliance teams should monitor whether the FTC files a test case against an enterprise deployer on agent nondisclosure, which would set a precedent on disclosure standards and harm attribution. The FTC Artificial Intelligence Compliance Plan remains the clearest existing framework for what the agency expects, and any update to it after these remarks would be significant. State-level enforcement is also a parallel risk: attorneys general in states with consumer protection mandates may act before the FTC moves formally. Watch also for any signal that Congress accelerates federal agent disclosure requirements through vehicles like the Protecting Consumers From Deceptive AI Act.