According to BeInCrypto, Gary Marcus has likened emerging AI agents to cars equipped with cardboard brakes, arguing that current development lacks essential safeguards. The comparison reflects concerns about reliability and safety in the deployment of autonomous AI systems capable of performing multi-step tasks.
At the same time, the U.S. Federal Trade Commission is opening a formal investigation into OpenAI and Anthropic, two major developers working on advanced AI agents. The probe focuses on practices related to AI development, testing, and market conduct, though specific details of the inquiry have not been publicly disclosed.
Marcus’s remarks emphasize that developers have yet to implement robust internal safeguards that could prevent harmful or unintended behavior from AI agents operating in real-world environments. He points to current limitations in oversight, verification, and accountability mechanisms as central issues requiring attention before broader deployment.
The FTC’s decision to open a probe signals increased regulatory scrutiny of leading AI firms amid growing public and expert concerns about safety, transparency, and responsible use of generative AI technologies. No statements from the agencies or companies were included in the source material.
