Treasury Secretary Scott Bessent told the House Financial Services Committee on Tuesday that AI developers should remain legally responsible for their software's outcomes. He rejected proposals for liability waivers, arguing such moves would allow a few powerful firms to dictate national policy.
Scott Bessent's Warning to the House Financial Services Committee
During his testimony on Tuesday, Treasury Secretary Scott Bessent argued that the U.S. government must resist granting liability exemptions to the most prominent artificial intelligence firms. according to the report, Bessent believes that allowing these developers to escape legal responsibility for the outputs of their technology would create a dangerous precedent of regulatory captture. He cautioned that such a shift would essentially allow a small group of private companies to shape national policy from behind the scenes.
Bessent maintained that the most secure approach to governance is to ensure that the creators of AI technology remain liable for any unwanted or harmful outcomes produced by their systems. By keeping the legal burden on the developers, the Treasury Secretary aims to prevent the emergence of what he described as a "slippery slope" leading toward a cartel of billionaires who could dictate a regulatory framework designed solely to favor their own interests.
Dario Amodei's 4 ,000-Word Plea for a "Frontier" Slowdown
The current tension stems from a proposal by Anthropic CEO Dario Amodei, who published a 4,000-word essay titled "We Must Pace the Frontier." In this piece, Amodei argued that the rapid pace of AI development is currently outstripping the government's ability to draft necessary laws. He suggested that industry groups should take the lead in setting safety standards, with the U.S. government acting as a mediator.
As reported, Amodei proposed a narrow waiver system that would allow companies to test new models without the immediate threat of liability, provided they remain accountable for catastrophic events, such as those causing death or serious injury to a large number of people. This vision of "coordinated slowdown" has found support among other industry titans, including OpenAI's Sam Altman, SpaceX's Elon Musk, and former DeepMind CEO Demis Hassabis.
From "Blackmail" to "Doomsday Narratives": The Sacks and Bedoya Critique
Not all observers view the industry's call for a slowdown as a benevolent move toward safety. David Sacks, who previously served as an AI czar for Donald Trump, characterized the proposal for industry-led standards and waivers as "blackmail." Similarly, former FTC commissioner Alvaro Bedoya warned that the industry might be leveraging a "doomsday narrative" regarding AI risks specifically to justify the creation of a protected cartel.
Adding another layer to the debate, FTC Chair Andrew Ferguson noted that OpenAI is already collaborating with Google and Anthropic on safety protocols. This existing cooperation led global policy chief Chris Lehane to claim that an antitrust waiver is unnecessary, as the firms are already coordinating their safety efforts without the need for formal legal immunity.
The Shadow of Section 230 and the Communications Decency Act
The current struggle over AI liability mirrors a historical precedent set during the rise of the social media era. Just last summer, various social media platforms relied on Section 230 of the Communications Decency Act to shield themselves from liability regarding content posted by their users. This legal framework allowed the early internet to scale rapidly by removing the threat of constant litigation over third-party speech.
Lawmakers are now grappling with whether AI developers should receive a similar shield or if the risks associated with generative models are too systemic to allow for such protections. The clash highlights a fundamental tension: while the tech industry argues that liability chills innovation, officials like Scott Bessent argue that public protection must take precedence over the speed of deployment.
Who Defines "Serious Harm" in the Amodei Proposal?
Despite the detailed nature of the debate, several critical points remain unaddressed in the current proposals. Specifically, the source does not clarify who would be responsible for defining the threshold of "serious harm" or what constitutes a "large number of people" in the context of Dario Amodei's proposed waivers. Furthermore, it remains unclear how the government would monitor the "narrow" testing phases to ensure that firms do not expand their immunity beyond the agreed-upon limits.
Comments 0