Artificial intelligence executives are pitching sweeping policy overhauls to shape how their technology gets regulated, but they may be vastly overestimating Washington's appetite for fundamental change.
OpenAI's Sam Altman and Anthropic's Dario Amodei have become the public faces of a push to reshape policy around AI development. OpenAI recently released an industrial policy paper drawing comparisons to the Industrial Revolution itself, proposing aggressive interventions including tax reform and a four-day workweek. Anthropic has focused on transparency measures, economic audits to track AI's effect on employment, stricter export controls, and government evaluation of AI systems.
The appeal of these proposals rests partly on the personalities behind them. OpenAI's chief global affairs officer, Chris Lehane, has spent decades as a political operative pushing for wealth redistribution tied to technological change. He previously championed a "new deal" for cryptocurrency. Anthropic's public policy head, Sarah Heck, came from Stripe and held roles in the Obama White House focused on global entrepreneurship and national security.
Yet history suggests Congress struggles with technology policy. The legislative branch has repeatedly stalled on privacy, social media regulation, and now artificial intelligence, often getting bogged down in technical details while facing intense industry lobbying and competing constituent pressures.
The State-Level Gambit
Recognizing this dysfunction, AI companies are shifting focus. Lehane told media outlets that OpenAI is concentrating its lobbying efforts at the state level, where political will moves faster especially in election years when voters demand assurances they will benefit from new technology. "Every politician says they lead, but what they typically do is they follow where the voters are," Lehane said.
Anthropic has backed state-level transparency bills while also calling for federal standards, betting that state-by-state victories can establish momentum toward broader change.
The tension between Silicon Valley and Washington runs deeper than legislative gridlock. The two regions operate on fundamentally different timescales and philosophies. One prioritizes speed and disruption; the other moves slowly or not at all. Nand Mulchandani, former chief technology officer at the CIA and the Pentagon's Joint Artificial Intelligence Center, noted the emerging power struggle. "Both coasts think that they're in charge," he said. "But Silicon Valley now has power rivaling the power of what a government has. What we're seeing now is the first large fight over who's driving the bus."
The Trump administration has demonstrated those limits repeatedly. Efforts to block state-level AI regulation at the federal level have failed, and the White House's recent AI framework proposal faces skepticism in Congress despite administration backing.
For AI companies, the current dynamic may not be entirely bad. They can advocate for transformative policies knowing the odds of passage are minimal, all while positioning themselves as responsible actors who warned Washington about the stakes. The gap between ambition and political reality has become the message itself.
Comments