Lawmakers urge AI leaders to stop warning of existential doom

Washington officials and members of Congress from both parties are urging AI executives to cease framing their technology as an existential threat, arguing that constant warnings of digital sentience and global catastrophe are turning a regulatory debate into a political liability. Sources close to the White House and Congress told Axios that lawmakers view the current crop of AI leaders as culturally strange and out of touch, preferring they focus on building products safely rather than centering public discourse on potential disasters.

Sen. John Curtis (R-Utah) described the industry as operating in a vacuum, stating that it is difficult to imagine a group more out of touch with everyday Americans. Rep. Ro Khanna (D-Calif.) expressed distrust in the ability of billionaires to make safety decisions, citing comments by OpenAI CEO Sam Altman about accepting negative outcomes for technological benefits. Sen. Bernie Moreno (R-Ohio) wrote to Anthropic CEO Dario Amodei, complaining about "panic-inducing commentary" that wreaks havoc on public discourse. Democrats are also leveraging these tensions, with Sen. Chris Murphy (D-Conn.) calling on his party to make the AI industry's political connections and data centers a central issue to protect workers.

Anthropic has drawn specific criticism from administration officials over recent safety incidents involving its models that affected government websites. The company attended a late September gathering with President Trump but did not disclose these incidents at the time. Instead, Anthropic informed U.S. officials on Thursday, Oct. 8, and published a blog post about the events on Friday evening. One official told Axios that the company was "hiding the ball." Anthropic stated it briefed the White House and affected agencies once its reviews were complete and is working with the Super Intelligence Force to develop an incident reporting process.

The company's legislative strategy has also generated confusion. Anthropic proposed an AI safety bill that included federal preemption of state AI laws, a stance the company has previously opposed. The bill defined a catastrophic incident as causing 50 deaths or more, a threshold that drew backlash. According to Politico, provisions defining AI catastrophe in terms of deaths have been part of landmark state regulations in California and elsewhere. Anthropic's Head of Public Affairs has since walked back parts of the bill, increasing distrust in Washington due to the shifting nature of the company's advocacy.

OpenAI faces similar scrutiny, with both companies targeted by a developing FTC investigation. New York State Assemblymember Alex Bores criticized OpenAI's policy proposals as vague, suggesting the company aims to block state regulation. In response to a blog post by OpenAI chief lobbyist Chris Lehane, an OpenAI spokesperson said the company is engaging with both parties, the administration, and statehouses to discuss the benefits of AI. However, a recent poll indicates that a majority of Americans still view AI as bad for society. Regulators remain uncertain whether the industry's frequent late-night news releases are deliberate attempts to manage narratives ahead of potential IPOs or the work of inexperienced teams navigating Washington.