facebook-pixel

Utah is studying AI in the wild today to help lawmakers write the rules for tomorrow

The state’s Office of AI Policy is letting companies test emerging technology under customized oversight while studying where the state might need more regulation.

(Chris Samuels | The Salt Lake Tribune) Utah lawmakers expect to set new regulations governing artificial intelligence.

(Chris Samuels | The Salt Lake Tribune) Utah lawmakers expect to set new regulations governing artificial intelligence.

Utah is experimenting with artificial intelligence to find the best ways to regulate it, including letting some companies test cutting-edge AI under customized state oversight.

The state’s Office of Artificial Intelligence Policy uses a learning lab and a “regulatory sandbox” — a controlled testing ground where businesses can safely launch new and untested products — to help develop policy recommendations for the Legislature.

Zach Boyd, who heads the office, said during an interim committee meeting this week that agreements with the companies let them “go forward and potentially beyond what current regulations specify” so the state can give lawmakers more information before they decide what rules are necessary.

(Bethany Baker | The Salt Lake Tribune) Zach Boyd, who oversees Utah's Office of Artificial Intelligence Policy, says the state has reached agreements with companies.

It is authorized to do so by the state’s Artificial Intelligence Policy Act, which is set to be repealed in July 2027 but now will move to a process where legislators periodically review whether it’s still effective.

Lawmakers established the office in 2024 with the regulatory sandbox and a learning lab that studies AI applications with companies, industry groups and academics.

The learning lab is meant to identify issues that could require changes in state policy. One of its projects resulted in HB452 during the 2025 legislative session, which created new requirements and consumer protections around AI in mental health care.

The office also helped inform legislation addressing AI-generated deepfakes during the 2026 session.

It’s now focused on AI agents — autonomous software systems that use an artificial intelligence model to plan, use tools, and execute multistep actions toward a specific goal with minimal human supervision — overall and in a medical setting.

There are lots of potential applications, Boyd said, including AI agents that could help gradually adjust prescription dosages.

The agency has received dozens of applications from leading companies, he said, that are looking to deploy clinical AI.

“We’re trying to figure out how to prudently clear some of those,” Boyd said, “so that we can still oversee them and learn from them.”

That approach reflects one of the state’s three objectives for the office, said Margaret Busse, who heads the Utah Department of Commerce. Those goals were fostering innovation, protecting the public, and proactively learning about a technology that was expected to grow rapidly.

Busse said the state wanted the lessons from the regulatory sandbox to directly inform recommendations to lawmakers.

“That’s, in fact, exactly what is happening right now with the healthcare and AI piece,” she said.

Boyd didn’t give any official policy recommendations but expects to in October and November.

Support free news for Utah

sltrib.com is now free to access — no subscription required. We made this decision because we believe access to trustworthy, independent news shouldn’t depend on what you can afford — especially as misinformation and AI-generated content continue to rise.

Free to read doesn’t mean free to produce. Our reporters show up every day to ask hard questions and hold powerful institutions to account. That work takes resources. As a nonprofit newsroom, we rely on support from people who believe it matters. Make a donation today to fund local news that serves Utah communities.

You can help us bring more local news to more communities today.