A federal judge on Monday issued a delay in enforcing Colorado’s artificial intelligence law, touted by backers as a tool to protect consumers and residents from algorithmic discrimination but which critics called heavy-handed and unworkable.

The lawsuit against Senate Bill 24-205 was filed by Elon Musk on behalf of his xAI company. The Trump administration joined in on the lawsuit last week.

The U.S. Justice Department pointed to the 2024 law’s “explicit carveout for discriminatory algorithms designed to advance ‘diversity’ or ‘redress historic discrimination.'”

“Laws that require AI companies to infect their products with woke DEI ideology are illegal,” said Assistant Attorney General Harmeet K. Dhillon of the Justice Department’s Civil Rights Division. “The Justice Department will not stand on the sidelines while states such as Colorado coerce our nation’s technological innovators into producing harmful products that advance a radical, far-left worldview at odds with the Constitution.”

In pushing for the legislation in 2024, backers said “algorithmic discrimination” is affecting decisions in hiring, housing applications, and health care coverage, and the legislation provides a framework to protect Colorado residents.

“AI systems are evolving faster than we can write and pass policy on them – which is why we need to act now,” Senate Majority Leader Robert Rodriguez, D-Denver, said in a statement two years ago. “Many system’s algorithms have biases baked in and can easily result in discriminatory outcomes when it comes to housing applications, hiring practices, and more.”

The legislation’s author added the bill would establish “foundational guardrails” but that the measure is “just a first step, and as technology continues to evolve, our work in this space must evolve alongside it.”

The motion delaying enforcement comes just as lawmakers are racing toward the end of the legislative session on May 13. That leaves policymakers just two weeks to make changes to the 2024 law before its June 30, 2026, implementation date.

Monday’s order from Magistrate Judge Cyrus Y. Chung of the U.S. District Court for Colorado said the defendant, Attorney General Phil Weiser, will not initiate enforcement for alleged violations of the 2024 law within 14 days of the court ruling on xAI’s motion for a preliminary injunction.

The delay in enforcement had been sought by both parties, according to the order issued Monday.

The April 24 motion seeking the delay also stated Weiser would not initiate enforcement against xAI for alleged violations of Colorado’s law.

xAI agreed to submit a motion for a preliminary injunction within 28 days after final adoption of rulemaking implementing SB 24-205 or any legislation that could replace or amend it.

The Center for Democracy & Technology, a Washington, D.C.-based organization, said Tuesday it is unfortunate “that once again Colorado officials have decided to delay protecting their constituents from the real and documented harms of AI in decisions affecting people’s lives and livelihoods.”

The statement noted that several groups are working on the 2024 law and argued that its fundamental tenets — which seek to prohibit alleged discrimination in housing, employment, or access to education or healthcare “through the use of opaque and unaccountable algorithmic systems” — must be maintained.

Two draft measures have already surfaced at the Capitol, one from a task force convened by the governor and the other from Rodriguez, the Denver Democrat who originally sponsored the AI law.

Under the Rodriguez draft obtained by Colorado Politics, the attorney general must adopt rules by Dec. 31, 2026.

The draft states the attorney general is directed to enforce the legislation through the “Colorado Consumer Protection Act” and a violation of the provisions of the bill is deemed a deceptive trade practice.

One sticking point already surfacing in the Rodriguez draft is that a developer or deployer — meaning an end user of AI technology, such as a school or small business — shall retain records, such as system identifiers, change logs, documentation, and notices of material updates, for three years.

The retention policy applies when a developer creates an automated decision-making technology (ADMT) with the intent that it be used to make “consequential decisions,” or if the developer becomes aware that the ADMT is being used to make those decisions consistent with its contracted uses.

Gov. Jared Polis, who expressed misgivings even as he signed the original law, has asked policymakers to address fears raised by the industry that, he said, have been “fueling critical technological advancement.” He also raised worries about Colorado “going it alone.”

The governor earlier convened a working group to “navigate appropriate and necessary changes” to the 2024 law, which, at the time of its enactment, was one of the first in the nation that regulated “algorithmic discrimination.”

It was originally set to take effect on Feb. 1, but lawmakers last year pushed it back to June 30, 2026 to give more time to reach an agreement with Big Tech, consumer advocates and business interests.

As policymakers worked on the legislation, Palantir Technologies decided to move its headquarters from Colorado to Florida, raising fresh worries about the state’s business climate, with a new report warning the departure could shave hundreds of millions of dollars from the state economy and signal broader challenges for its technology sector.

The artificial intelligence and software company, one of Colorado’s most valuable publicly traded firms, had announced moving its headquarters to a suburb of Miami after five years in Denver.

While offering few details about the decision, Palantir had expressed several worries about Colorado in its annual reports filed with the Securities and Exchange Commission. These reports often detail any risks that could affect the company or its investors.

Palantir cited the new AI regulations as a cause of concern, according to its annual reports for the fiscal years 2024 and 2025.

Palantir compared Colorado’s “state-level oversight” to the European Union’s Artificial Intelligence Act.

“Compliance with such obligations may be difficult, onerous, and costly, and could adversely affect our business, reputation, financial condition, results of operations, and growth prospects,” the company said in its report filed earlier this year.