The Michigan Department of Insurance and Financial Services has issued new guidance outlining expectations for how financial services providers, including insurers, use artificial intelligence.
The bulletin sets a clear direction. AI adoption now comes with formal governance strings attached.
Under the guidance, regulated entities are expected to design, implement, and maintain a written AI Systems Program, referred to as an AIS Program.
The framework aims to reduce the risk of negative consumer outcomes as AI tools move deeper into underwriting, claims, pricing, and customer interaction.
The bulletin recognises AI’s upside. It points to product innovation, improved consumer interfaces, process automation, and operational efficiency.
It also draws a hard line around risk. Inaccuracy, unfair discrimination, data exposure, weak transparency, and opaque decision logic sit high on the concern list.
According to Beinsure analysts, the emphasis reflects a broader regulatory shift away from principle-only guidance toward enforceable process expectations, even where formal AI rules remain limited.
DIFS sets out a broad framework for how AIS Programs should be constructed, placing responsibility squarely on governance, risk management controls, and internal audit oversight. The expectation is clear.
AI supervision must sit inside core organisational processes rather than operate as a standalone or experimental function.
According to Beinsure analysts, regulators are signalling that AI risk belongs at board and executive level, not buried within technical teams or outsourced vendors.
Anita Fox, director of DIFS, said artificial intelligence continues to reshape financial services, raising compliance expectations alongside opportunity.
She said firms must ensure AI systems comply with federal and state law while keeping consumer protection front and centre.
