Health Insurance - Artificial Intelligence, Adverse Decisions, and Grievances - Reporting Requirements
HB697 would expand reporting requirements for Maryland health insurance carriers by adding a new quarterly report to the Maryland Insurance Commissioner on the creation, deployment, and use of artificial intelligence and automated decision-making systems. Carriers would have to disclose when and why these systems are used, who trains them, what data and methods are used, what guidance they rely on, and what bias-testing and mitigation steps are taken. The bill is aimed at increasing transparency around how AI is used in health insurance operations, especially where automated tools may affect coverage-related decisions.
The bill also amends existing quarterly reporting rules for carrier grievance and adverse decision data. It adds new reporting items such as the number of grievances filed, the monetary value of grievance decisions, average hold and total call times for grievance and appeal centers, and the monetary value of adverse decisions outside the subtitle’s grievance process. Existing reporting on adverse decisions, grievances, emergency cases, prior authorization, step therapy, and related outcomes would remain, but with more detailed aggregation and additional financial and operational metrics. The Commissioner would continue to compile an annual summary report for the Governor and General Assembly.
HB697 would amend the Insurance Article to impose new quarterly disclosure obligations on health insurers operating in Maryland, particularly regarding AI and automated decision-making systems and expanded grievance/adverse decision reporting. It would not directly regulate the use of AI itself, but it would require carriers to document and report their AI practices and decision-making processes to the Maryland Insurance Commissioner, creating a stronger oversight record for the state. The bill would also broaden the data carriers must provide on grievances, appeals, call center performance, and the monetary value of decisions, which could affect carrier compliance systems and reporting workflows.
Based on the bill text and available context, the measure appears to be framed as a transparency and oversight bill rather than a punitive one, and it is sponsored by a large bipartisan-looking group of delegates. There are no committee transcripts or recorded votes provided, so no formal debate or vote-based sentiment can be identified from the available record. The overall tone of the bill suggests concern about the growing use of AI in insurance decision-making and a desire for more public and regulatory visibility into carrier practices.
The likely points of contention are the added compliance burden on health insurance carriers and the breadth of information required about proprietary AI systems, training methods, data sources, and bias testing. Carriers may view the reporting requirements as administratively burdensome or as exposing sensitive operational information, while supporters are likely to argue that the disclosures are necessary to monitor fairness, detect bias, and understand how automated tools affect coverage and grievance outcomes. Another possible area of concern is the inclusion of monetary-value reporting for grievances and adverse decisions, which may raise questions about how those values are calculated and used by regulators.