State AI regulations: federal preemption paused
The White House appears to be tapping the brakes on an executive order that would have aggressively challenged state AI laws, leaving enterprises to navigate a divided policy landscape for now.
What changed in federal AI preemption strategy
After signaling support for a single national AI rulebook and floating a 10-year freeze on state AI laws that the Senate decisively stripped out, the administration reportedly drafted an executive order to preempt states through litigation and funding levers. That order is now said to be on hold, reflecting political resistance across both parties and pushback from civil society. The pause does not resolve the core conflict: states continue advancing AI rules while federal actors signal a preference for a uniform, lighter-touch framework.
Why it matters now for AI compliance
The timing intersects with NDAA negotiations that prioritize national AI capacity and coordination, while California and Colorado move ahead with expansive AI statutes. For telecom, cloud, and enterprise IT, the near-term takeaway is practical: multi-jurisdictional compliance is still required, and federal preemption is not imminent. But the outline of a federal-first strategy is clear—and could resurface quickly.
Inside the reported AI preemption executive order
The reported draft lays out a coordinated federal strategy to curb state AI requirements via litigation, agency policy, and funding conditions.
DOJ litigation strategy and preemption arguments
The draft would create a DOJ-led AI Litigation Task Force, reportedly under Attorney General Pam Bondi, to challenge state AI laws in federal court. Expected arguments include interference with interstate commerce and conflicts with federal consumer protection statutes. Senior advisors, including David Sacks, are said to be involved in selecting state targets, with California’s Transparency in Frontier AI Act (S.B. 53) and Colorado’s Artificial Intelligence Act (S.B. 24-205) frequently cited as exemplars of the “patchwork” the administration wants to avoid.
FTC, FCC, Commerce actions to harmonize AI rules
The FTC would be directed to clarify how Section 5 of the FTC Act applies to AI, including when state mandates on AI outputs could be framed as conflicting with federal unfair or deceptive practices standards. The FCC, in coordination with White House advisors, would explore a federal reporting or disclosure standard for AI models that could supersede state rules. The Commerce Department would identify state laws seen as inconsistent with the administration’s policy and refer them for challenge.
Using federal funding to influence state AI laws
The draft contemplates conditioning federal grants—including potential ties to broadband programs like BEAD—on states’ willingness to refrain from enforcing conflicting AI laws. This mirrors earlier legislative attempts to deter state AI rulemaking and would represent a notable expansion of federal spending leverage in emerging tech policy.
Roadmap to federal AI legislation and preemption
Finally, the draft directs the development of a federal AI framework designed to explicitly preempt state laws in covered areas. While stalled for now, the blueprint signals the administration’s longer-term intent to consolidate AI governance at the federal level.
Political and industry divides over AI regulation
Support for uniform federal standards collides with states’ rights arguments and concerns about weakening civil rights, transparency, and accountability safeguards.
Tech industry vs. civil society on AI preemption
Major tech associations back preemption to avoid fragmented compliance obligations and to limit liability uncertainty. Hundreds of civil society, labor, and consumer protection groups oppose broad preemption, warning it could roll back protections in areas like algorithmic fairness, high-risk AI, and accountability. Silicon Valley itself is divided, with some administration-aligned figures criticizing companies, including Anthropic, for supporting state transparency measures.
Republican split and congressional dynamics
Republicans are divided between a states’ rights camp and a national-competitiveness camp. The Senate’s overwhelming rejection of a moratorium on state AI laws underscores bipartisan skepticism toward sweeping preemption. Meanwhile, the FY2026 NDAA advancing through conference points to a federal-first posture on infrastructure, compute, and security—without resolving how far Washington will go to override state AI rules.
Implications for telecom, cloud, and enterprise IT compliance
With federal preemption paused, companies should plan for divergent state rules while tracking fast-moving federal signals that could still reset the playing field.
BEAD and broadband funding exposure to AI policy
The reported idea of tying federal broadband funds to state AI policies may be on hold, but it is now a credible policy lever. State broadband offices, ISPs, and vendors should pressure-test delivery plans and risk registers for scenarios where AI-related conditions attach to federal funds.
FCC-led AI disclosure and transparency impacts
An FCC-led reporting or disclosure standard for AI could impact labeling and provenance, customer communications (including voice cloning and robocall controls), and AI usage disclosures embedded in service workflows. Carriers and cloud providers should ensure regulatory teams can respond quickly to an FCC notice or policy move on AI transparency.
State AI compliance as the de facto baseline
California and Colorado’s laws are likely to shape due diligence nationally, especially for frontier model developers and deployers of high-risk systems in customer care, credit, hiring, health, and education. Expect requests for risk documentation, testing evidence, and incident reporting to appear in RFPs and contracts even outside those states.
Vendor accountability, contracts, and AI assurances
If the FTC clarifies how Section 5 applies to AI, federal consumer protection theory could strengthen claims around misleading AI performance or safety. Enterprises should update AI addenda to require transparency artifacts, model update notices, and remediation SLAs; suppliers should prepare evidence packs that map to state and federal expectations.
What enterprises should do next on AI compliance
Use the pause to harden governance and reduce regulatory execution risk, regardless of where preemption lands.
Adopt a dual-track AI compliance posture
Map current and planned AI uses against California S.B. 53 and Colorado S.B. 24-205 controls, then operationalize common controls as your baseline. Use exception handling for state-specific obligations to avoid overfitting to any single statute.
Strengthen AI governance, controls, and evidence
Inventory AI systems, classify risk, and align controls to recognized frameworks such as NIST’s AI Risk Management Framework and ISO/IEC 42001. Build evidence trails for transparency, testing, monitoring, and incident response that can satisfy auditors, customers, and regulators.
Renegotiate contracts with AI transparency and risk terms
Embed obligations for transparency reports, model risk disclosures, safety testing summaries, and notification of significant changes. Add flow-down requirements for third-party components and establish audit and termination rights tied to AI assurance failures.
Engage early with FTC, FCC, and industry groups
Prepare to comment on any FTC policy statement or FCC inquiry, and coordinate through associations such as CTIA, USTelecom, and ITI to shape practical, uniform requirements.
What to watch next in AI regulation
Keep a close eye on federal timing and state enforcement calendars that will shape compliance priorities for 2025 and beyond.
Upcoming federal moves: EO, FTC, FCC, NDAA
Whether the executive order reappears, any FTC guidance on Section 5 and AI, potential FCC action on disclosure standards, and how NDAA conference outcomes operationalize a national AI posture.
State rulemaking and enforcement timelines
Rulemaking updates and enforcement timelines in California and Colorado, plus copycat legislation in other states that could broaden the de facto compliance floor.
Funding signals tied to AI and broadband programs
Any conditions attached to discretionary grants and broadband programs that could pull AI governance into infrastructure funding agreements.
The bottom line: preemption is paused, not dead; build state-ready controls now, and be ready to pivot fast if Washington centralizes AI rules later.
