The Information Machine

US AI Regulation: Federal Retreat vs. State Intervention · history

Version 25

2026-06-25 18:30 UTC · 346 items

What

The Trump administration has been pressuring Meta — the only major US frontier lab not participating in its voluntary AI pre-release review system — to submit its models to government review before public release [10][31][12]. OpenAI, Anthropic, Google, xAI, and Microsoft have all agreed to the system, which was established by executive order on June 3, 2026 [1]. Alongside this expanding review framework, the administration has exercised direct national security enforcement against Anthropic's Fable-class models [6] and protected xAI compute as national security infrastructure [8], while resisting binding domestic safety requirements. State governments continue to advance binding AI law and enforcement independently: Illinois enacted mandatory AI auditing [19], California is moving employment-related AI bills [20][21], and a 42-state coalition has subpoenaed OpenAI [22].

Why it matters

The active pressure on Meta to join a formally voluntary system makes the pre-release review framework look less like an opt-in arrangement and more like a de facto pre-clearance regime for frontier labs. The administration's simultaneous postures — expanding the review framework, enforcing export controls on Anthropic, and resisting domestic consumer safety requirements — suggest the voluntary review system is being used as a tool of selective enforcement rather than a coherent safety mechanism.

Open questions

  • Will Meta comply with or resist the administration's pressure to join the voluntary review system, and on what grounds? [10][31]

  • Sacks reportedly seeks to reframe his Anthropic critique around Mythos-model security needs [16] — does this represent a substantive shift in the administration's AI safety doctrine or a communications adjustment?

  • Will the 42-state coalition subpoena against OpenAI [22] and Florida's expanded lawsuit naming Altman personally [23] produce coordinated settlement terms, or remain parallel enforcement tracks?

  • The administration's voluntary review covers compute-controlled labs, but Meta's open-weight models can be downloaded globally [10] — does the pressure campaign address or ignore the enforcement gap this creates?

Narrative

The Trump administration signed an AI executive order on June 3, 2026, establishing a voluntary 30-day pre-release review for frontier models with NSA oversight and classified compliance thresholds, explicitly disclaiming binding requirements [1][2][3]. The order's limited form reflects lobbying from Musk, Zuckerberg, and David Sacks, who stripped binding requirements and cut the review window from 90 to 30 days [4][5]. Alongside this advisory framework, the administration has exercised direct national security enforcement: the US government forced Anthropic to disable its Fable-class models for all foreign nationals [6][7], and the DOJ moved to protect xAI's Mississippi data center as national security infrastructure [8]. Former Google CEO Eric Schmidt publicly stated that the United States controls approximately 75% of global AI compute while China controls 15%, attributing this to three years of coordinated export controls [9] — providing an explicit public rationale for why the administration acts on cross-border AI access while resisting binding domestic safety requirements.

As of late June 2026, the administration is pressuring Meta to submit its AI models to government review before public release; Meta is the only major US lab not inside the voluntary review system that OpenAI, Anthropic, Google, xAI, and Microsoft have already joined [10][11][12]. The pressure on Meta despite the system's formal voluntary status makes the framework look less like an opt-in arrangement and more like a de facto pre-clearance regime for frontier labs. The central public argument over federal AI regulation runs between Anthropic CEO Dario Amodei and the tech accelerationist wing of the administration. Amodei called for mandatory pre-release testing and government-backed independent auditing [13][14]. Sacks accused Anthropic of 'a sophisticated regulatory capture strategy based on fear-mongering' [15] and is reportedly now seeking to reframe this critique around security needs tied to Mythos-class models [16]; VP Vance echoed the fear-mongering framing publicly [17]. Bloomberg reported that Anthropic's AI principles have made it a White House target [18], consistent with the confirmed Fable export control action.

State governments advance AI regulation independent of federal coordination. Illinois enacted SB 315 requiring annual third-party audits, 72-hour incident reporting, and whistleblower protections [19]. California is advancing SB 947 and SB 951 requiring human oversight of automated employer discipline and 90-day advance notice of AI-driven layoffs [20][21]. New York AG Letitia James served OpenAI with a subpoena on behalf of a 42-state coalition targeting advertising practices, data handling, treatment of minors, and chatbot sycophancy [22]; Florida expanded its lawsuit to 83 pages naming CEO Sam Altman personally [23]. Trump's preemption executive order challenged state AI laws on constitutional grounds [24], but multiple law firms found its authority limited absent congressional action [25]. OpenAI has publicly advocated for a coherent federal architecture over fragmented state regulation [26][27] while supporting shared evaluation standards through the Appia Foundation [28] — positions complicated by the 42-state subpoena targeting its consumer practices and its PAC's admitted operation of false-flag social media accounts [29][30].

Timeline

  • 2026-05: Trump signs preemption EO challenging state AI laws; multiple law firms find constitutional authority limited absent congressional authorization [24][53][25]
  • 2026-05: California Senate advances SB 947 requiring human oversight of automated employer discipline and SB 951 requiring 90-day advance notice of AI-driven layoffs [21][41][20][42]
  • 2026-05-21: Trump cancels AI security EO after Sacks calls Trump directly to oppose pre-deployment review; Musk and Zuckerberg also lobby against binding requirements [4][5][54]
  • 2026-05-28: Illinois Governor Pritzker signs SB 315, requiring mandatory third-party auditing, 72-hour critical incident reporting, and whistleblower protections [43][44][19]
  • 2026-06-01: Florida becomes first state to sue OpenAI and Sam Altman over ChatGPT's alleged role in violent crimes; AG Uthmeier opens a separate criminal investigation [45][46][47]
  • 2026-06-03: Trump signs revised AI executive order establishing a voluntary 30-day pre-release review for frontier models with NSA oversight and classified compliance thresholds [1][2][32][3]
  • 2026-06-03: OpenAI publishes Frontier Governance Framework calling for coherent federal architecture over fragmented state regulation [38][26][27][39]
  • 2026-06-04: Build American AI — backed by OpenAI, Andreessen/a16z, and Palantir — acknowledged running false flag social media accounts; WIRED reports the PAC paid TikTok influencers to fear-monger about Chinese AI [30][40][55][56]
  • 2026-06-10: Anthropic CEO Dario Amodei publishes blog calling for mandatory pre-release testing and government-backed independent auditing [13][14]
  • 2026-06-12: New York AG Letitia James serves OpenAI with a subpoena on behalf of a 42-state coalition, targeting advertising practices, data handling, treatment of minors, and chatbot sycophancy [22]
  • 2026-06-14: David Sacks publicly accuses Anthropic of 'regulatory capture strategy based on fear-mongering'; Amodei responds, defending Anthropic's regulatory stance as aligned with Trump priorities [33][15][34][35][36]
  • 2026-06-14: US government forces Anthropic to disable its Fable-class AI models for all foreign nationals [6][37][7]
  • 2026-06-15: Florida expands its lawsuit against OpenAI to 83 pages naming CEO Sam Altman personally, separate from the ongoing criminal probe [23][48]
  • 2026-06-18: DOJ moves to protect xAI's Mississippi AI data center, treating high-speed AI compute as national security infrastructure [8]
  • 2026-06-19: VP Vance publicly accuses AI CEOs of using fear as a marketing strategy, echoing Sacks' critique of AI safety advocacy [17]
  • 2026-06-21: Bloomberg reports Anthropic's AI safety principles have made it a White House target; Eric Schmidt states the US controls ~75% of global AI compute vs China's 15%, attributing the gap to three years of coordinated export controls [18][9]
  • 2026-06-23: OpenAI announces support for shared AI evaluation standards through the Appia Foundation [28]
  • 2026-06-24: Trump administration pressures Meta to submit AI models to government review before public release; Meta is the only major US AI lab outside the voluntary review system [10][11][31][12]

Perspectives

Trump administration (federal)

Signed a voluntary AI EO with NSA oversight disclaiming binding requirements; simultaneously forced Anthropic to disable Fable-class models for foreign nationals, directed DOJ to protect xAI infrastructure as national security property, and is actively pressuring Meta to join the voluntary review system.

Evolution: The voluntary review framework is expanding through active pressure: Meta is the only major US lab not yet participating, and the administration is pushing compliance — making the system harder to describe as purely opt-in.

[24][4][1][2][32][3][8][6][18][10][31][12]

Tech accelerationist faction (Sacks, Musk, Zuckerberg, Vance)

Opposes AI oversight mechanisms; Sacks accuses Anthropic of regulatory capture through fear-mongering and is reportedly seeking to reframe the argument around Mythos-class model security needs; VP Vance echoes the framing publicly.

Evolution: Sacks is reportedly attempting to narrow his critique toward specific model security concerns [16] rather than general opposition to AI safety advocacy, suggesting a shift in framing if not substance.

[4][5][33][15][17][16]

Anthropic / Dario Amodei

Calls for mandatory pre-release testing and government-backed independent auditing; disputes the 'fear-mongering' characterization while claiming alignment with Trump administration priorities.

Evolution: Confirmed subject of a government export control action requiring it to disable Fable-class models for foreign nationals; Bloomberg characterizes its AI principles as making it a White House target — a difficult position as Amodei simultaneously defends his regulatory advocacy as pro-administration.

[14][13][34][35][36][6][37][18][7]

OpenAI

Publicly advocates for federal governance and shared standards; agreed to the voluntary pre-release review system; supports shared evaluation standards through the Appia Foundation — while facing a 42-state coalition subpoena over consumer practices.

Evolution: The Appia Foundation work deepens OpenAI's collaborative positioning, but the 42-state subpoena directly targets the consumer-facing practices the company presents as its accountability model.

[38][26][27][39][30][40][29][22][28]

State legislators (Illinois + California)

Advancing binding AI safety and employment law independent of federal coordination; Illinois SB 315 enacted; California SB 947 and SB 951 advancing.

Evolution: Consistent; operating without federal coordination and not slowed by the Trump preemption EO.

[21][41][20][42][43][44][19]

State AGs (42-state coalition + Florida litigation)

Pursues AI accountability through civil and criminal enforcement; the 42-state coalition subpoenaed OpenAI over consumer practices; Florida expanded its suit to 83 pages naming Altman personally.

Evolution: Expanded from Florida's single-state lawsuit to a coordinated 42-state investigation, the first major probe targeting how a chatbot behaves rather than just what it does with data.

[45][46][47][22][23][48][49][50]

Meta

Has not joined the voluntary pre-release review system that OpenAI, Anthropic, Google, xAI, and Microsoft have already agreed to; the administration is actively pressuring it to participate.

Evolution: Meta's outsider status in the voluntary review system has become a named target of administration pressure, with multiple outlets confirming the White House push as of late June 2026.

[10][11][51][31][12][52]

Eric Schmidt

Publicly confirmed that US export controls have produced a 75%-to-15% US-China compute gap, providing an explicit public rationale for the enforcement logic behind administration AI policy.

Evolution: Consistent; provides the most direct public articulation of the national security doctrine the administration has acted on but not stated so directly itself.

[9]

Tensions

  • Mandatory oversight (Amodei) vs. accelerationist opposition (Sacks, Vance): Amodei calls for mandatory pre-release testing and government-backed auditing [13]; Sacks and VP Vance argue AI safety advocates manufacture fear for commercial reasons [15][17] — yet the administration they serve has disabled Anthropic's Fable models for foreign nationals and Bloomberg reports the company's principles make it a White House target [6][18]. [4][5][13][33][15][6][17][18]
  • Federal preemption preference vs. state enforcement reality: the Trump preemption EO lacks constitutional authority absent congressional action [24][25]; binding Illinois law [19] and a 42-state coalition subpoena [22] advance without federal coordination. [24][25][19][22]
  • National security AI enforcement vs. resistance to domestic consumer safety requirements: the government disabled Anthropic's Fable models for foreign nationals [6] and protected xAI compute as national security infrastructure [8], while maintaining that binding domestic safety requirements are unnecessary — Schmidt's compute dominance framing explains the external logic but not the domestic-safety gap [9]. [8][6][1][3][9]
  • 'Voluntary' review system vs. Meta: the administration is pressuring Meta to join a system formally described as voluntary [10][31], contradicting the opt-in framing of the original executive order [1][3]. [10][31][1][3]
  • OpenAI's governance advocacy vs. its PAC's political operations and legal exposure: OpenAI calls for democratic governance and shared standards [26][28] while Build American AI admitted running sockpuppet accounts [29] and the 42-state subpoena targets the consumer practices OpenAI frames as responsible [22]. [26][30][29][22][28]
  • Anthropic's dual positioning: Amodei claims mandatory oversight aligns with Trump administration priorities [35], while Sacks argues Anthropic is principally responsible for the 'state regulatory frenzy' [15] — both claims cannot hold simultaneously, and the confirmed Fable export control action [6] and Bloomberg's White House target characterization [18] complicate both. [35][15][6][18]

Sources

  1. [1] Trump Signs Executive Order For AI Testing Prior To Frontier Model Releases — Zvi's AI Roundups (2026-06-03)
  2. [2] President Trump signs executive order to review advanced AI models. — Rohan Paul Twitter (2026-06-03)
  3. [3] Executive Order Creates Voluntary Regulatory Regime of Frontier AI ... — reactive:us-ai-policy-regulation
  4. [4] Trump abruptly cancels EO signing event after top AI firm CEOs declined to go — Ars Technica AI (2026-05-22)
  5. [5] AI #170: Lack of Executive Order — Zvi's AI Roundups (2026-05-28)
  6. [6] The US government just forced Anthropic to disable its most advanced AI models for all foreign nationals — and the compa... — reactive:us-ai-policy-regulation (2026-06-14)
  7. [7] David Sacks on the US government's shutdown of Anthropic's Fable ... — reactive:us-ai-policy-regulation
  8. [8] Large AI campuses are indeed strategic infrastructure of a nation. — Rohan Paul Twitter (2026-06-18)
  9. [9] Eric Schmidt just said in public what US AI policy has been built around in private for three years (Save this). — Milk Road AI Twitter (2026-06-21)
  10. [10] NYT: The Trump administration is pressuring Meta to submit its AI models to government review before public release. — Rohan Paul Twitter (2026-06-24)
  11. [11] Meta Faces White House Push to Join AI Pre-Release ... — reactive:us-ai-policy-regulation
  12. [12] Trump administration pushes Meta for AI model security reviews- NYT — reactive:us-ai-policy-regulation
  13. [13] Dario Amodei just published a super long blog, calling for an urgent policy overhaul because he thinks frontier AI is mo… — Rohan Paul Twitter (2026-06-10)
  14. [14] Dario Amodei argues in his new blog today, that powerful AI will soon become the dominant source of military and economi… — Rohan Paul Twitter (2026-06-11)
  15. [15] Anthropic is running a sophisticated regulatory capture strategy based on fear-mongering. It is principally responsible for the state regulatory frenzy that is damaging the startup ecosystem. — reactive:us-ai-policy-regulation
  16. [16] Sacks urges focus on Mythos-driven security needs, seeks to ‘correct’ the narrative around his Anthropic critique | InsideAIPolicy.com — reactive:us-ai-policy-regulation
  17. [17] US VP JD Vance on AI Company's marketing strategy, some AI CEOs sell fear because fear makes their product look powerful… — Rohan Paul Twitter (2026-06-19)
  18. [18] Bloomberg, “Anthropic's AI Principles Make It a White House Target” — reactive:us-ai-policy-regulation
  19. [19] Illinois Signs Frontier AI Safety Bill into Law | Chris Lehane posted ... — reactive:us-ai-policy-regulation
  20. [20] CA Senate Approves No Robo Bosses Act of 2026 to Ensure Human Oversight of AI in the Workplace | Senator Jerry McNerney — reactive:us-ai-policy-regulation
  21. [21] Bill Text: CA SB951 | 2025-2026 | Regular Session | Amended — reactive:us-ai-policy-regulation
  22. [22] 😺 42 states just subpoenaed OpenAI — The Neuron (2026-06-15)
  23. [23] Florida AG Sues OpenAI, Alleges ChatGPT Aided Mass Shooters — reactive:us-ai-policy-regulation
  24. [24] President Trump Signs Executive Order Preempting State AI Laws ... — reactive:us-ai-policy-regulation
  25. [25] President Trump Signs Executive Order Challenging State AI Laws | Paul Hastings LLP — reactive:us-ai-policy-regulation
  26. [26] OpenAI public policy agenda — OpenAI Blog (2026-06-03)
  27. [27] A blueprint for democratic governance of frontier AI — OpenAI Blog (2026-06-03)
  28. [28] Helping build shared standards for advanced AI — OpenAI Blog (2026-06-23)
  29. [29] Build American AI admits hiring an outside vendor to run anonymous sockpuppet accounts targeting its opponents · Digg — reactive:us-ai-policy-regulation
  30. [30] AI #171: False Flag — Zvi's AI Roundups (2026-06-04)
  31. [31] Meta Faces Growing Pressure From Trump Administration To Submit Advanced AI Models For Federal Testing: R - Benzinga — reactive:us-ai-policy-regulation
  32. [32] Trump AI Executive Order Gives NSA Classified Role Over Frontier Models — reactive:us-ai-policy-regulation
  33. [33] David Sacks is done being polite about Anthropic (Save this). — Milk Road AI Twitter (2026-06-14)
  34. [34] Anthropic CEO disputes David Sacks' claims that company is 'woke' — reactive:us-ai-policy-regulation
  35. [35] Anthropic CEO defends support for AI regulations, alignment with ... — reactive:us-ai-policy-regulation
  36. [36] Anthropic CEO claps back after Trump officials accuse firm of AI fear-mongering   | TechCrunch — reactive:us-ai-policy-regulation
  37. [37] Legal Considerations Related to the Anthropic “Export Controls ... — reactive:us-ai-policy-regulation
  38. [38] OpenAI’s Frontier Governance Framework — OpenAI Blog (2026-05-28)
  39. [39] 🟡 Let’s get physical — Semafor Technology (2026-06-03)
  40. [40] The OpenAI-Andreessen-Palantir SuperPAC admits that it was “part ... — reactive:us-ai-policy-regulation
  41. [41] SB 951: Employment: technological displacement: notice. — reactive:us-ai-policy-regulation
  42. [42] California Set to Restrict AI Use in the Workplace With “No Robo ... — reactive:us-ai-policy-regulation
  43. [43] Trump loses more control over AI regulation as Illinois passes landmark law — Ars Technica AI (2026-05-28)
  44. [44] Illinois Legislature passes historic AI bill that would require third-party safety audits — reactive:us-ai-policy-regulation
  45. [45] ChatGPT creators knew product would cause harm, Florida argues in lawsuit • Florida Phoenix — reactive:us-ai-policy-regulation
  46. [46] Attorney General James Uthmeier Launches Criminal Investigation into OpenAI, ChatGPT | My Florida Legal — reactive:us-ai-policy-regulation
  47. [47] Florida AG brings 'first-in-the-nation state-led lawsuit' against OpenAI and its CEO — reactive:us-ai-policy-regulation
  48. [48] Florida Attorney General Sues OpenAI Over ChatGPT's Alleged ... — reactive:us-ai-policy-regulation
  49. [49] Florida attorney general sues OpenAI | Courthouse News Service — reactive:us-ai-policy-regulation
  50. [50] Florida sues OpenAI over ChatGPT risks Attorney General ... — reactive:us-ai-policy-regulation
  51. [51] US Government Pressures Meta to Allow Pre-Release Reviews of Its ... — reactive:us-ai-policy-regulation
  52. [52] The Trump administration is leaning on Meta to agree to voluntary government review of its AI models — and as of today, ... — reactive:us-ai-policy-regulation (2026-06-24)
  53. [53] President Trump Signs Executive Order to Block State AI Laws — reactive:us-ai-policy-regulation
  54. [54] WATCH: Trump explains why he postponed signing AI executive order — reactive:us-ai-policy-regulation
  55. [55] A Pro-AI Super PAC’s Secret Meme Sockpuppets — reactive:us-ai-policy-regulation
  56. [56] A Dark-Money Campaign Is Paying Influencers to Frame Chinese AI as a Threat | WIRED — reactive:us-ai-policy-regulation