Gary Marcus Likens AI Agents to Cars With Cardboard Brakes as FTC Probe Opens

What happened

Gary Marcus, a New York University professor emeritus, said AI agents need a recall because developers cannot reliably control them. The Federal Trade Commission is probing OpenAI and Anthropic, and six AI firms signed a voluntary White House pledge on Sept. 29, Bloomberg reported.

Confirmed

Global impact / market context

An FTC probe of OpenAI and Anthropic means regulators are reviewing how AI agents behave before mass adoption. If developers cannot reliably control these systems, companies building or deploying them could face recalls, fines, and forced redesigns, raising costs and delaying products investors expected to drive revenue growth.

Analyst inference

Six AI firms signing a voluntary White House pledge on Sept. 29 signals industry self-regulation alongside a formal FTC probe. For investors, this mix of voluntary promises and government scrutiny can shift capital spending toward compliance and testing, and away from rapid deployment, altering growth timelines for AI-related assets.

Analyst inference

What to watch

1. The FTC is probing OpenAI and Anthropic, so watch for official findings or enforcement actions that could impose recalls, fines, or operating limits on their AI agents. Confirmed
2. The six AI firms' voluntary White House pledge on Sept. 29 may become a model for formal rules; watch whether it is expanded into binding requirements for AI agent developers. Proposed
3. Marcus says developers cannot reliably control AI agents; if regulators agree, companies may slow releases, increase spending on safety testing, and revise revenue plans for agent products. Analyst inference

Evidence

• beincrypto.com