Treasury Secretary Rejects New Liability Protections for AI
Scott Bessent says AI firms should manage their own risks rather than seeking government-provided safe harbor.
Updated on Oct. 21, 2026 in Artificial Intelligence
Live Poll
Should the government provide liability protection to AI firms that choose to slow their development?
Treasury Secretary Scott Bessent has rejected industry claims that systemic artificial intelligence risks necessitate new liability protections. Bessent signaled that firms seeking to slow development to manage outcomes will not receive special legal shields from the Treasury.
Why it matters
Bessent characterizes existential warnings from AI industry leaders as alarmist, choosing to prioritize market accountability over government intervention. This stance indicates that the U.S. government is unlikely to implement a heavy regulatory response in the near term.
The Treasury is currently pursuing bilateral talks with China to establish communication channels for flagging potential AI-related cyberattacks or runaway systems. No specific technical parameters for these incident notifications have been formalized.
The players
Scott Bessent
Scott Bessent is the United States Treasury Secretary responsible for shaping the administration's economic and regulatory policy.
The details
Bessent emphasized that AI companies retain the power to halt operations independently to manage responsibility, rather than relying on the government for protection. He is simultaneously coordinating with international partners to secure global channels for reporting severe AI system failures.
Timeline
September 15, 2026: Bessent testified before Congress regarding AI regulation.
Early October 2026: Bessent discussed AI liability risks in interviews with Axios and CNBC.
October 21, 2026: Bessent addressed an investor conference regarding systemic AI risks.
The Tech Race
This decision marks a significant shift in how the government approaches the evolution of AI relative to the Communications Decency Act Section 230 safe harbor protections. The Treasury's refusal to extend liability shields marks a departure from historical safe harbor precedents applied to earlier internet technologies.
Because companies will hold full legal responsibility for product outcomes, users may see more cautious AI release cycles and rigorous safety testing. This policy forces firms to internalize the risks of their products, which could increase operational costs for the developers.
The takeaway
The government is shifting the burden of AI safety squarely onto the shoulders of the developers themselves. Industry participants should anticipate continued legal exposure rather than government-backed safeguards in the near term.
Further reading
For broader insight into the developing regulatory environment, visit the Artificial Intelligence section.
Live Poll
Should the government provide liability protection to AI firms that choose to slow their development?
