Artificial intelligence has rapidly moved from experimental labs into mainstream business operations, shaping industries as diverse as healthcare, finance, education, and logistics. As companies integrate machine learning and generative AI into products and services, state governments are beginning to confront the question of how to regulate these technologies in a way that protects consumers while still allowing innovation. In Michigan, this conversation has taken on a concrete form with the introduction of House Bill 4668 (HB 4668), a proposed act that would establish business standards for AI developers.
Unlike general debates about federal oversight or global treaties, HB 4668 is focused on the state level, making Michigan one of the first U.S. jurisdictions to consider specific legislative guardrails for AI development. If passed, this law would reshape how businesses build, train, and deploy artificial intelligence systems within the state. Understanding what is at stake requires a close look at the language of the bill, its intended protections, the obligations it places on developers, and the broader implications for businesses and consumers alike.
The push for legislation like HB 4668 comes from a combination of public pressure, economic ambition, and social responsibility. Michigan has long been a hub of innovation, historically leading in automotive manufacturing and now looking to position itself as a leader in advanced technologies. As AI becomes integrated into connected vehicles, manufacturing automation, and workforce applications, lawmakers see both an opportunity and a risk.
The risks are not abstract. From biased hiring algorithms to flawed facial recognition systems, AI technologies have already demonstrated their capacity to cause real harm. High-profile examples, such as wrongful arrests tied to algorithmic errors or discriminatory loan approval systems, have pushed policymakers to act. At the same time, Michigan leaders understand that unchecked regulation could stifle the very innovation they hope to attract. HB 4668 is presented as a middle ground an attempt to create standards that hold developers accountable while ensuring that businesses remain competitive in a rapidly changing technological environment.
At its core, HB 4668 outlines the responsibilities of businesses that develop and deploy artificial intelligence systems. It frames these responsibilities not merely as ethical guidelines but as enforceable business standards, giving the state authority to hold companies accountable. The bill requires developers to adopt clear documentation practices, to conduct testing that ensures their systems meet reliability and fairness thresholds, and to provide transparency about how their models make decisions.
A key feature of the bill is its emphasis on explainability. Lawmakers recognize that one of the most common criticisms of AI is its “black box” nature decisions are made, but the process behind them is opaque. HB 4668 would require businesses to create mechanisms for explaining how AI outputs are generated, particularly in cases where those outputs affect consumer rights or access to services. This does not mean revealing proprietary code, but it does mean providing intelligible explanations to regulators, auditors, or end users when necessary.
The consumer protection dimension of HB 4668 is among its most ambitious elements. The bill envisions a future where individuals affected by AI systems can demand accountability in the same way they can when harmed by a defective product. By requiring developers to monitor and audit their systems regularly, HB 4668 attempts to ensure that harmful biases are not simply discovered after the fact but are systematically reduced through proactive oversight.
The act also touches on issues of consent and notification. When a consumer interacts with an AI system whether it be through a chatbot, a hiring platform, or an automated decision-making tool the bill would require businesses to disclose that an algorithm is being used. This transparency is intended to restore a sense of agency to consumers, allowing them to understand when their data is being processed by AI rather than a human. If enacted, these provisions would mark a significant step toward embedding consumer rights into the fabric of AI development.
For businesses, the passage of HB 4668 would mean a shift in priorities. Until now, many companies have treated responsible AI practices as voluntary, guided more by public relations concerns than legal requirements. HB 4668 would change this dynamic by placing compliance at the center of AI development. Companies would need to create internal processes for documenting training data, validating system outputs, and reporting adverse outcomes.
One of the most significant obligations is the requirement for regular third-party audits. Just as financial statements are subject to independent verification, AI systems would need to be reviewed by external experts to confirm that they meet fairness and safety standards. This could impose new costs on businesses, particularly startups, but lawmakers argue that the investment will pay dividends by creating more trustworthy products and reducing legal risks down the line. For Michigan companies, compliance with HB 4668 could become a badge of credibility, signaling to partners and customers that they adhere to some of the nation’s first state-level AI standards.
Not surprisingly, HB 4668 has sparked debate among stakeholders. Proponents argue that the bill is essential to prevent abuse and ensure that Michigan residents are not left vulnerable to poorly designed AI systems. They point to examples of algorithmic harm in other states as evidence that waiting for federal regulation is not an option. By acting now, Michigan can both protect its residents and position itself as a national leader in ethical AI development.
Critics, however, warn that the bill could have unintended consequences. Some business leaders argue that requiring extensive documentation and third-party audits will slow down innovation and drive companies to states with less restrictive frameworks. Others worry that the law’s broad definitions of AI could unintentionally capture smaller businesses or tools not meant to be regulated at such a scale. The challenge, as always, lies in finding the balance between necessary oversight and flexibility for innovation.
If HB 4668 is enacted, Michigan could become a case study in state-level AI governance. Other states are watching closely, as the federal government has yet to pass comprehensive AI legislation. This creates a patchwork system where states experiment with different approaches, and Michigan’s choices could shape national conversations. For businesses operating across state lines, compliance may become increasingly complex, requiring them to adapt to multiple legal environments.
At the same time, the introduction of HB 4668 signals that the era of unregulated AI is coming to an end. Whether or not this specific bill passes, the message is clear: lawmakers are no longer content to let the market govern itself. The implications reach beyond compliance, touching on corporate reputation, consumer trust, and the ethical responsibilities of technology companies in the 21st century.
Looking forward, HB 4668 has the potential to reshape how AI is built in Michigan. Developers may need to integrate legal and ethical compliance into every stage of the design process, from data collection to deployment. This could foster closer collaboration between engineers, legal experts, and ethicists, creating multidisciplinary teams that approach AI not only as a technical challenge but as a societal responsibility.
For consumers, this may result in more reliable and transparent AI systems, giving them confidence that their data is being handled responsibly. For businesses, compliance may initially feel burdensome, but it could ultimately drive competitive advantage by ensuring that products are not only cutting-edge but also trustworthy. As Michigan lawmakers deliberate, the broader AI community will be watching, recognizing that what happens in one state could eventually influence the trajectory of AI governance across the nation.
House Bill 4668 represents more than just a local legislative proposal. It is part of a larger reckoning with the power and risks of artificial intelligence. By setting standards for AI developers, the bill seeks to embed accountability into the DNA of innovation, ensuring that technological progress does not come at the expense of human rights or consumer protection.
For businesses in Michigan, the message is clear: the regulatory landscape for AI is evolving rapidly, and preparation is essential. Those who adapt early will be best positioned to thrive in an environment where compliance and trustworthiness are as important as technical excellence. For consumers, HB 4668 offers hope that the benefits of AI can be realized without sacrificing fairness, transparency, or safety. And for the broader national conversation, Michigan’s proposed act may become a template for how states can lead in shaping the future of artificial intelligence.
Contact Tishkoff
Tishkoff PLC specializes in business law and litigation. For inquiries, contact us at www.tish.law/contact/. & check out Tishkoff PLC’s Website (www.tish.law/), eBooks (www.tish.law/e-books), Blogs (www.tish.law/blog) and References (www.tish.law/resources).
Sources
- Michigan Legislature, House Bill 4668, Proposed 2025 Session Documents https://www.legislature.mi.gov/Bills/Bill?ObjectName=2025-HB-4668
- Artificial Intelligence Governance and Regulation Report, Center for Strategic and International Studies https://www.csis.org/programs/wadhwani-ai-center/research-themes/governance-regulation
- “Algorithmic Accountability in the United States,” Brookings Institution Analysis, 2024 https://www.brookings.edu/articles/how-different-states-are-approaching-ai/
- “The Risks of AI Bias and Discrimination in Consumer Applications,” Harvard Business Review, 2024 https://hbr.org/2025/05/how-companies-can-navigate-the-age-of-ai-driven-inequality