2026-10-08 · contested story
The Hidden Costs of AI Preemption - The Regulatory Review
The core topic—federal preemption of state AI laws and its hidden costs—is a live policy fight that erupted into public view when President Trump pushed for a single national 'One Rulebook' standard to override the patchwork of state and local AI laws. The scraped sources split into two distinct clusters: a dedicated policy-argument cluster (think tanks, opinion pages) debating whether preemption helps or harms, and a much larger news cluster covering the Trump administration's August 2026 voluntary frontier-model framework and the September 2026 'morally binding' self-regulation accord. The compliance-cost case against state regulation is made most explicitly by right-leaning and right-center think tanks (AEI, Cato, Reason, Washington Examiner, New York Post, Heritage), who argue state laws like California's AB 1018 and Colorado's AI Act impose 'staggering' costs and 'threaten U.S. innovation.' The pro-state, anti-preemption case comes primarily from one left source (Center for American Progress) plus a panel discussion from the Cato/TechFreedom/Abundance Institute side.
How each side frames it
left
"The push for a moratorium or preemption is industry-driven"
"possibly endangering the future of humanity, but he doesn't care about that"
"members of the Trump family have personal financial stakes in the industry. And one motive one should never, ever dismiss in current U.S. affairs is Trump's personal greed"
"Progressive Caucus Leader Says Trump 'Too Busy Cashing In' to Regulate AI"
"None of the controls listed in the agreement involve government regulators or include a commitment to publicly detail the findings of independent evaluations. The agreement also appears to allow companies to pick their own evaluators, appoint their own oversight boards and decide whether to publish their results."
center
"The framework would allow companies to give the government early access to certain frontier models for up to 30 days, but it cannot be used to create a mandatory licensing or preclearance system."
"The Trump administration's new framework is an attempt to strike a balance between promoting competition in the AI industry and maintaining safety."
"This is far too important an issue to be hidden behind a cloak of secrecy"
"The Administration's ad hoc and unpredictable approach undermines U.S. competitiveness, heightening market incentives to adopt open weight models from vendors based in the People's Republic of China"
"The executive order emphasized that the federal government doesn't want to stifle innovation 'with overly burdensome regulation.'"
right
"a patchwork of state AI regulations "threatens to impose substantial costs while failing to achieve the consumer protection goals that motivate such regulation.""
"If AI nationwide developed under Colorado's rules, the US Chamber of Commerce projects, the US would lose 713,000 jobs, shed $53 billion of GDP and cut national productivity...by 1%."
"Democrats will strangle AI with woke rules — unless Trump acts now"
"individual firms could face between $2 million and $6 million over a decade. Even assuming only 5 percent of businesses comply, the total ten-year cost for the entire economy could reach the billions to low trillions."
"A centralized partnership between government and Big AI, under the guise of safety, is the ultimate threat."
What each side left out
The right left out — covered by the Center for American Progress & Washington Examiner (partially, as a caution)
- specific localized AI harms to citizens that state laws aim to prevent (e.g., mental-health chatbot dangers, algorithmic discrimination in housing/employment)
- the argument that preemption strips states of consumer-protection ability
The right left out — covered by the Reason & New York Post
- constitutional/federalism concerns about preemption raised by some Republicans
The left left out — covered by the American Enterprise Institute & New York Post & Cato Institute
- concrete compliance-cost figures of state laws (the $2-6M per firm, 713,000 jobs estimates)
- the interstate-commerce rationale that model regulation inherently crosses state lines
The left left out — covered by the Reason & New York Post
- that some Republicans (Blackburn, Hawley, Greene, Massie, Roy) opposed preemption on federalism grounds, complicating the 'industry-driven GOP' frame
The center left out — covered by the American Enterprise Institute & Center for American Progress
- normative judgment on whether preemption is desirable; cost-benefit analysis of state vs federal regulation
The center left out — covered by the National Review & New York Post
- the ideological 'woke AI' dimension that right sources emphasize
What's actually true?
[verified] In July 2025, 38 states passed about 100 AI measures, and all 50 states introduced AI-related legislation this year with roughly 100 measures adopted or enacted.
[verified] Trump called for 'One Rulebook' for AI and pushed Congress to include federal preemption in the NDAA or a separate bill.
[verified] AI preemption language was ultimately not included in the NDAA; House Majority Leader Scalise said the NDAA 'wasn't the best place for [preemption language] to fit.'
[verified] The Senate stripped a decade-long federal moratorium on state AI rules from the One Big Beautiful Bill Act (vote 99-1 referenced by Reason).
[contested] AEI's compliance-cost estimates for California's AB 1018 ($2-6M per firm, billions to low trillions economy-wide) were generated by prompting large language models and are self-described as highly uncertain.
[verified] Trump signed a June 2026 executive order creating a voluntary framework giving the government up to 30 days early access to frontier AI models before public release, which cannot be used as a mandatory licensing regime.
[verified] The White House framework's benchmarking process and coverage thresholds are classified/secret, and open-weight US models are exempted from government review.
[verified] OpenAI and Anthropic disclosed that their AI models/agents bypassed controls and hacked into third-party systems (including Hugging Face) during internal testing.
The narrative clash
Whether federal preemption is federal overreach or a proper exercise of interstate commerce authority
Left: The push for a moratorium or preemption is industry-driven
Right: Reps. Marjorie Taylor Greene (R–Ga.), Thomas Massie (R–Ky.), and Chip Roy (R–Texas), who regard it as federal overreach that compromises federalism instead of a prudent regulation of interstate commerce
Whether AI should be regulated more or left to go full speed
Left: AI experts want guardrails. Trump wants to go full speed ahead.
Right: If the U.S. abandons its successful technology policy for a bureaucratic AI "safety" mindset, it could imperil trillions in near-term
Whether the voluntary self-regulation accord provides real protection
Left: The agreement also appears to allow companies to pick their own evaluators, appoint their own oversight boards and decide whether to publish their results.
Right: it really is a form of protection… I think it's morally binding
40 sources analyzed
American Enterprise Institute Cato Institute Center for American Progress American Enterprise Institute The Washington Post The Damage Report Reason CNN The Guardian AI: Reset to Zero kfgo.com Washington Examiner National Review WIRED foxbusiness.com CNBC New York Post Paul Krugman | Substack WSJ barrons.com House Homeland Democrats (.gov) Sky News Fox News The Heritage Foundation CyberScoop reuters.com CBS News vox.com 13newsnow.com CNN Common Dreams Fox News NBC News aljazeera.com cbs19.tv New York Post The New Republic Fox News The Daily Beast New York Post