The integration of AI in U.S. hospitals has led to a $942 million increase in costs for insurers, according to a BCBSA report.
The Blue Cross Blue Shield Association (BCBSA) has reported that the integration of artificial intelligence (AI) in U.S. hospitals has led to a staggering increase of $942 million in costs for insurers. This rise is attributed to the enhanced capability of AI tools to identify more diagnoses, which subsequently elevates service bills despite no changes in the quality of care.
AI is increasingly utilized in medical coding processes, where it assesses physician notes and lab results to uncover additional comorbidities. This has resulted in a higher classification of hospitalizations as medically complex, which directly impacts reimbursement rates for hospitals.
Analyzing insurance claims from early 2023 to late 2025, BCBSA found that the percentage of hospital admissions deemed medically complex grew from 37% to 40%. This shift caused insurers to incur an additional $942 million in costs over this timeframe, with around 70% of that amount related to secondary diagnoses. In more than 55,000 cases, these added diagnoses raised claim complexity, resulting in an extra $653 million disbursed to hospitals.
“It is critically important that we did not find any changes in the care that would correspond to a more complex patient,” explained Luca Choker, senior vice president at BCBSA and co-author of the report.
Choker emphasized that this cost estimate does not account for cases where hospitals documented additional services, suggesting that the observed increase in costs may reflect altered documentation practices rather than actual changes in patient care.
Additionally, the report highlighted discrepancies in diagnosis rates versus treatment levels. For instance, in hospitals where cases of acute post-hemorrhagic anemia were frequently reported, blood transfusions occurred less often, indicating that increased diagnoses may not correlate with worsening patient conditions.
“One should ask: why is it that one hospital, treating similar patients under similar conditions, starts to deviate?” noted Razia Hashmi, BCBSA's vice president for clinical affairs, suggesting that technology may exaggerate these discrepancies.
Despite rising concerns, the report did not confirm a direct link between specific AI technologies and the resultant cost increases. It also acknowledged that while insurance claims were used for analysis, comprehensive clinical records would offer deeper insights into patients' conditions.
Counterarguments from Hospitals
The American Hospital Association (AHA) has countered claims that AI is solely responsible for the rising costs, asserting that hospitals are encountering a higher volume of patients with more severe conditions, driven by an aging population and increasing chronic diseases.
According to AHA, data from Vizient indicated a 5% rise in the complexity index of hospital patients between 2019 and 2024, supporting their view that AI assists in accurately capturing existing health conditions rather than creating unnecessary complexity. In this debate, both hospitals and insurers utilize automation: hospitals for billing optimization and insurers for claim checks.
Recent statements by healthcare authorities, including Mehmet Oz from the Centers for Medicare and Medicaid Services, acknowledged that while AI may initially inflate healthcare expenses, its efficiency could eventually streamline existing billing processes. Research at the Mayo Clinic has shown promising advancements in AI detection of diseases, such as pancreatic cancer, potentially aiding earlier diagnosis and treatment.
Overall, while AI is revolutionizing diagnostics and coding in healthcare, its financial implications and the ongoing debate about its impact on care quality continue to unfold.
