Utah is betting that artificial intelligence will only thrive if public trust comes first, using targeted laws, real penalties, and supervised experimentation to prove that safety and innovation don’t have to be opposites. (Source: Image by RR)

Lawmakers Introduce Real Penalties to Enforce Responsible AI Development

Utah has emerged as an unlikely national leader in artificial intelligence governance, advancing a regulatory strategy that aims to balance innovation with public trust. At the center of this effort is Margaret Woolley Busse, executive director of Utah’s Department of Commerce, who describes AI as facing a growing “political crisis.” Drawing lessons from years of litigation against social media companies, state leaders believe unchecked AI risks repeating the same extractive data practices that fueled public backlash against Big Tech.

Rather than pursuing sweeping federal-style mandates, Utah’s approach operates on two tracks. The first includes targeted legislation under the Pro-Human AI Initiative, such as laws addressing deepfakes and AI companions. The second is a more ambitious proposal—the Artificial Intelligence Transparency Act—which would require developers of frontier AI models to publish child safety plans, protect whistleblowers, and face civil penalties of up to $3 million for violations. These efforts, according to an article in ksl.com, signal a willingness to regulate with real enforcement power while still leaving room for innovation.

Utah has also created a regulatory “Learning Lab” that allows AI companies to experiment under strict oversight through customized Regulatory Mitigation Agreements. One of the most prominent examples is Doctronic, a state-approved pilot program that allows AI to assist with prescription renewals under phased human supervision. The system cannot prescribe controlled substances and requires doctors to manually validate hundreds of early cases, ensuring AI augments clinical decision-making rather than replacing physicians outright.

State officials argue this supervised model represents a “third way” between laissez-faire tech acceleration and heavy-handed regulation. By pairing transparency, accountability, and revocable permissions with room to experiment, Utah hopes to build public confidence in AI systems before fear turns into resistance. For Busse and her team, trust is not a byproduct of innovation—it is the prerequisite for AI’s long-term survival in society.

read more at ksl.com