In a decision that’s stirring national conversation, Virginia Governor Glenn Youngkin recently vetoed a bill aimed at regulating the use of artificial intelligence in high-stakes decision-making. The bill, known as the High-Risk Artificial Intelligence Developer and Deployer Act (HB 2094), had passed the state legislature with support from bipartisan lawmakers and consumer advocates. It would have been one of the most comprehensive AI consumer protection laws in the United States.
Instead, its rejection leaves Virginians—and Americans more broadly—asking an urgent question: Can innovation coexist with ethical guardrails in AI?
HB 2094 sought to address a growing concern in AI: algorithmic bias in decisions that affect people’s lives. The bill specifically targeted what it called “consequential decisions,” including:
- Employment and hiring
- Housing access
- Financial services and lending
- Educational opportunities
- Parole and criminal justice outcomes
- Healthcare eligibility and coverage
It required developers and deployers of “high-risk AI systems” to assess their algorithms for potential discrimination and report their practices transparently. If passed, it would have set a precedent for AI accountability nationwide, following in the footsteps of Colorado’s AI Act passed in 2024.
Delegate Michelle Lopes Maldonado, who sponsored the bill, described it as a “flexible and breathable framework”—designed to adapt as technology evolves without hamstringing innovation.
Governor Youngkin disagreed with the bill’s approach. In his veto statement, he described the legislation as imposing a “burdensome artificial intelligence regulatory framework” that could stifle innovation, especially among startups and smaller companies unable to meet its compliance demands.
Instead, Youngkin pointed to existing anti-discrimination laws and his own 2023 Executive Order establishing AI governance within the executive branch. He argued that these measures already address the risks the bill sought to mitigate.
In essence, the veto reflects a larger national debate: How much regulation is too much when it comes to AI?
Although the bill didn’t focus exclusively on healthcare, its implications are deeply relevant to the field. From diagnostic algorithms to AI-driven insurance decisions, healthcare is one of the most consequential arenas for algorithmic decision-making.
Already, studies show troubling evidence of bias in health-related algorithms:
- A 2019 study published in Science revealed that an algorithm used by hospitals to manage care for millions of patients systematically underestimated the health needs of Black patients.
- A 2023 Brookings report warned that health insurance algorithms could deny care based on flawed or biased data inputs, especially for patients with complex or chronic conditions.
In such a landscape, regulatory frameworks like HB 2094 could serve as critical safeguards—ensuring that innovation doesn’t come at the expense of fairness or trust.
Virginia is not alone in grappling with this issue. Here’s a quick look at what’s happening elsewhere:
- Colorado: Enacted a comprehensive AI law in 2024 that mandates regular bias audits and transparency reporting for high-risk systems.
- California: Currently considering a similar bill with input from Silicon Valley leaders, patient advocates, and civil rights groups.
- New York: Implemented regulations around AI in hiring, requiring employers to disclose the use of AI tools and their decision-making criteria.
The takeaway? AI governance is coming—it’s just a matter of how, and when.
Virginia’s veto highlights the delicate balancing act between protecting the public and fostering innovation. Advocates argue that without regulation, AI systems can perpetuate existing inequities, especially in sectors like healthcare, housing, and employment. Critics worry that overly rigid rules could slow down progress or drive innovation out of state.
Both sides raise valid points. But perhaps the real opportunity lies in co-designing policy with technologists, ethicists, and communities alike. It’s not a binary choice—it’s a challenge to craft regulation that’s both adaptable and accountable.
As AI continues to shape healthcare, justice, finance, and beyond, it’s clear we need rules—not just code—to build a future we can trust.
