Hospital Group Challenges Blue Cross’s Claim Linking AI to Higher Medical Bills

Higher hospital bills do not necessarily mean AI is inventing medical problems. In responses published by Healthcare Brew on October 7, the American Hospital Association and a health-tech adviser challenged Blue Cross Blue Shield’s explanation for rising billing complexity. The dispute centers on whether AI is inflating claims or finding conditions that previously went undocumented.

The responses address a September 24, 2026 report from the Blue Cross Blue Shield Association, or BCBSA. It estimates that increased billing complexity added $942 million in spending for the Blues system between 2023 and 2025. The association attributes some of the increase to hospitals using AI to help enter patient-care codes.

More diagnoses, higher reimbursement

The billing mechanism starts with the medical record. AI tools can scan records and transcribe patient encounters, surfacing secondary diagnoses beyond the main reason for hospitalization. Those diagnoses can affect how much an insurer pays because hospital reimbursement reflects the conditions documented and the complexity assigned to a case.

BCBSA’s report says the share of cases billed as medically complex rose from roughly 37% in early 2023 to about 40% by late 2025. Its analysis covered 55,000 cases classified as excess complex cases. The spending estimate applies to the Blues system, not the entire healthcare industry.

The association alleges hospitals added secondary diagnoses that moved claims into higher-severity, higher-paying categories without corresponding increases in clinical care. It says much of the increase came from upcoding at teaching hospitals—billing cases at a higher level of complexity.

It can’t tell the difference between a diagnosis that was inappropriately added and one that was always present but previously missed.

Missy Harbert, solution advisor at health-tech company Revecore

The missing distinction in billing data

Harbert’s objection is methodological: the analysis uses insurance claims rather than a review of patients’ medical charts. She accepts that unsupported codes should not be defended and says accuracy must govern both sides of a claim. But she disputes whether this data can separate improper additions from better documentation of genuine conditions.

American Hospital Association spokesperson Ben Teicher offered a different explanation for the trend. He told Healthcare Brew that patients are older and more clinically complex, and that AI can help providers capture their conditions for care planning. He said BCBSA’s analysis lacks the context needed to assess effects on quality, patient access or spending.

Blue Cross wants care to justify the codes

BCBSA’s response rests on treatment, not just documentation. David Merritt, its senior vice president of external affairs, said genuinely greater medical complexity would bring corresponding increases in treatment intensity and resource use. That is the association’s reason for questioning whether more complex claims reflect sicker patients.

Merritt called for guardrails rather than slowing innovation: a flexible, risk-based approach to AI, appropriate human involvement and accountability measures. His proposed standard is that coding and reimbursement accurately reflect patient care. The provider-side challenge is whether claims data alone can show when that standard has been breached.