xAI LLC v. Weiser (No. 1:26-cv-01515, D. Colo., filed April 9, 2026) is a federal constitutional challenge brought by xAI LLC against the enforcement of the Colorado AI Act (SB 24-205, codified as the Colorado AI provisions, CPAI). The complaint advances six counts under the First Amendment, the dormant Commerce Clause, the Due Process Clause, and the Equal Protection Clause. The U.S. Department of Justice moved to intervene on the plaintiff's side on April 24, 2026, and the court stayed enforcement of the Act while the challenge proceeds.
Infobox
| Field | Value |
|---|---|
| Case number | 1:26-cv-01515 (D. Colo.) |
| Parties | xAI LLC (plaintiff); U.S. Department of Justice (intervenor, April 24, 2026); Philip J. Weiser, Colorado Attorney General (defendant) |
| Court | U.S. District Court, District of Colorado |
| Filed | April 9, 2026 |
| Status | Active; enforcement of the Colorado AI Act stayed |
Parties
The plaintiff is xAI LLC, a Nevada corporation with its principal place of business in Palo Alto, California, and the developer of the Grok family of frontier models. The defendant is Philip J. Weiser, the Colorado Attorney General, sued in his official capacity as the sole enforcement authority under the Colorado AI Act. The U.S. Department of Justice entered the case as an intervenor on the plaintiff's side on April 24, 2026 (see Procedural history).
Claims
The complaint sets out six constitutional counts.
First Amendment. Three counts proceed under the First Amendment. The first, framed as free speech and compelled editorial modification, contends that training-data curation, system-prompt drafting, fine-tuning, and guardrail design are expressive conduct, and that CPAI's reasonable-care duty compels redesign to match a state-preferred fairness framework. The second, viewpoint discrimination, contends that CPAI's carve-out for differential treatment "intended to increase diversity or redress historical discrimination" creates asymmetric viewpoint regulation, invoking 303 Creative v. Elenis (2023) and Moody v. NetChoice (2024). The third, compelled disclosure, contends that training-data summaries, known-limitations disclosures, and impact-assessment reporting are content-based compelled commercial speech, and that the Zauderer "purely factual and uncontroversial" safe harbor fails given an "inherently subjective" discrimination evaluation.
Dormant Commerce Clause. The fourth count contends that CPAI reaches any developer "doing business in this state" and attaches liability for outputs affecting Colorado residents regardless of where produced, which the complaint frames as extraterritorial overreach under Healy v. Beer Institute and a Pike-balancing failure given the risk of contradictory regimes.
Due Process. The fifth count, vagueness, contends that the terms "reasonable care," "algorithmic discrimination," "historical discrimination," "substantial factor," and "material legal or similarly significant effect" supply no ascertainable standard, and that the statute "invites arbitrary enforcement."
Equal Protection. The sixth count contends that the diversity and historical-discrimination carve-out is a race-conscious classification that must survive strict scrutiny under Students for Fair Admissions v. Harvard (2023), and cannot.
Commentary has described the suit as the first full constitutional challenge to a state AI anti-discrimination statute. Observers have noted that a ruling upholding First Amendment shields for training-data and system-prompt choices could bear on Colorado, on the EU AI Act's extraterritorial provisions, and on the National Retail Federation's New York algorithmic-pricing challenge (New York Algorithmic Pricing Regime and AG Enforcement); that the Equal Protection count seeks to extend SFFA v. Harvard into AI governance; and that a federal-preemption count rooted in Executive Order 14365 — Ensuring a National Policy Framework for AI has been described as plausible but is not consistently reported in coverage.
Procedural history
xAI filed the complaint on April 9, 2026.
On April 24, 2026, the U.S. Department of Justice filed a motion to intervene, joining xAI's effort to block enforcement of the Colorado AI Act. DOJ argued that the law violates the Constitution by requiring discrimination, a framing that parallels the SFFA v. Harvard posture of the Equal Protection count and aligns the federal executive branch with the private plaintiff rather than acting through preemption rule-making. Coverage characterized the move as a federal-state confrontation over a state AI civil-rights statute likely to shape how other state AI anti-discrimination statutes, including those of Illinois and New York, are litigated (Source: bloomberg.com; Source: insideaipolicy.com). DOJ Assistant Attorney General Harmeet K. Dhillon stated in announcing the intervention that laws requiring AI companies to "infect their products with woke DEI ideology are illegal," a framing that aligns DOJ with the vocabulary of the "Preventing Woke AI" EO 14319 and treats CPAI's reasonable-care duty as ideologically coercive (Source: justice.gov). UCLA law professor Andrew Selbst characterized the DOJ intervention on April 27, 2026 as extending the Harvard affirmative-action ruling (Students for Fair Admissions v. Harvard) into AI, describing the federal posture as treating CPAI's diversity carve-out as the doctrinal heir to race-conscious admissions (Source: politico.com).
On April 27, 2026, the federal judge granted a joint motion to stay enforcement of the Colorado law, filed jointly by DOJ and xAI pursuant to the DOJ's April 24 motion to intervene. According to Bloomberg Law on May 1, 2026, the stay's procedural footprint rests on Equal Protection grounds, with DOJ's intervention arguing that the requirement to test for and mitigate disparate impact violates the Fourteenth Amendment Equal Protection Clause (Source: news.bloomberglaw.com). Following the DOJ intervention and the court order, Colorado state officials agreed on April 28, 2026 to a temporary stay on enforcement of the Colorado AI Act, suspending CPAI's scheduled compliance deadlines while the constitutional challenge proceeds. The Center for Democracy and Technology criticized the stay as an unjustified pre-merits suspension of a duly enacted civil-rights statute (Source: insideaipolicy.com).
Current status
The case is active, with enforcement of the Colorado AI Act stayed pending resolution of the constitutional challenge.
Note on confidence
Confidence is rated medium because the primary PACER complaint text was not retrieved. Count-by-count specifics are triangulated from JURIST, The Hill, PPC Land, NetChoice, FIRE, and Reclaim the Net, and should be verified before citing specific count structures.
Source summary
Full source page: xAI LLC v. Weiser — Complaint (D. Colo. 1:26-cv-01515) (sources/).
Relationships
- contradicts: Colorado AI Act (SB 24-205) and SB 25B-004 (Date Amendment) as a regulatory regime; tests AI and the First Amendment boundary.
- depends-on: AI and the First Amendment, AI Compliance Industry / Regulatory Fragmentation.
- related: Executive Order 14365 — Ensuring a National Policy Framework for AI (federal-preemption framing), New York Algorithmic Pricing Regime and AG Enforcement (parallel First Amendment NRF challenge), xAI, Grok (xAI), (Source: washingtonpost.com) (contextual coverage).