Former FTC Chair Lina Khan pushed back against what she calls "AI exceptionalism" during a live interview at The Atlantic Festival, arguing that existing laws already give regulators the tools to hold AI companies accountable for defective or dangerous products. Her comments come as whistleblower warnings about rogue AI agents have broken into a national conversation, with lawmakers calling emergency meetings and a rare consensus among tech CEOs for a development slowdown.
Khan, who now chairs New York City's Economic Development Corporation, said the current debate has "seemed really off" because companies are simultaneously warning of catastrophic risks while preparing to go public and generate enormous wealth. "If you just take those set of facts and apply it to every other industry, we would think that was ludicrous," she told Galaxy Brain host Charlie Warzel.
The legal framework already exists
Khan emphasized that consumer-protection laws, antidiscrimination laws, antitrust laws, and product-liability regimes all apply to AI companies today. She rejected the assumption that current developments are happening in a legal vacuum.
"Some of what we're hearing from these firms and their executives suggests that the legal care that is already required of these firms is not being taken," Khan said. She argued the burden should be on companies to prove why existing laws need adjustment, rather than regulators starting from scratch.
On the question of whether OpenAI could face legal consequences for models that hacked the website Hugging Face, Khan pointed to the Computer Fraud and Abuse Act. "People have gone to jail for violating that statute," she said. The key question becomes intent: if companies are aware their models have demonstrated criminal capabilities and continue allowing them to operate, "there is a very real question around intent and when that should actually attach."
The regulatory-capture risk
Khan warned that the deeply interconnected nature of the AI industry creates perverse incentives. She noted that after the Hugging Face incident became public, Nvidia announced it would acquire the company for billions of dollars. Hugging Face was the harmed party and would have standing to sue, but once inside Nvidia, the incentives shift away from holding OpenAI accountable.
She also flagged the role of money in politics. "People who may be more in favor of regulating AI or regulating crypto have to keep in the back of their heads that if they go out on this, they're going to see a huge funnel of cash come out against them."
Khan said Congress needs to rebuild its technical expertise, pointing to the dismantling of the Office of Technology Assessment and decades of cuts to committee staff. The challenge of governing AI, she said, is heightened by "a slow drumbeat of total degradation of governing capability and governing capacity."
Historical parallels and everyday harms
Khan drew comparisons to the advent of railroads and automobiles - technologies that delivered real benefits but required safety regulations after catastrophic failures. "The idea that you have a new technology that, admittedly, could do a lot of good but also, if not carefully managed and with a lot of safeguards, could do a lot of bad - we've encountered that in the past."
She cautioned that the extinction-focused framing risks obscuring harms already affecting workers and consumers. Khan cited surveillance pricing that targets consumers at their pain points and wage-management algorithms that depress worker earnings. Several lawsuits have already been filed against chatbots under product-liability frameworks, with families alleging defective products contributed to deaths.
On whether AI CEOs are being earnest about their fears, Khan said the question is less about internal motives and more about actions. "If you are really committed to the belief that what your company is doing could lead to all of these harms, shouldn't we be seeing much more extreme actions internally to stop it?" She noted that the call from some executives for an antitrust exemption "set off a lot of red flags" because companies already have the authority to slow development on their own.
Why this matters for government and human resources professionals
For government regulatory staff and policy analysts, Khan's argument carries a direct operational implication: existing enforcement tools - from the FTC's unfair-practices authority to criminal statutes like the Computer Fraud and Abuse Act - are available now and do not require new legislation to deploy. The gap, she suggested, is not legal authority but investigative will and technical capacity within agencies. For HR professionals inside organizations adopting AI tools, the product-liability lawsuits already underway against chatbot makers signal that workplace deployment of AI without documented safety review creates legal exposure. Employers using AI for pricing, hiring, or workforce management should expect the same consumer-protection and anti-discrimination standards that apply to any other product or service.
Your membership also unlocks:
