Blue Cross study links AI documentation tools to $1B rise in insurer costs
A Blue Cross Blue Shield Association study found that AI-assisted documentation and coding added nearly $1 billion in costs for insurers over two years. Providers used AI to scan records and draft notes, leading to more complex patient coding and higher reimbursements. The analysis attributed $653 million to secondary diagnoses and $942 million in total added costs compared with 2023.
The study examined inpatient billing data from Blue Cross companies covering more than 100 million people, focusing on hospital admissions. Among major bowel surgery patients, secondary condition diagnoses rose sharply—partial intestinal blockages increased 55 percent and excess acid diagnoses grew 33 percent between early 2023 and late 2025. These additional diagnoses did not correspond with increased treatments like blood transfusions for anemia patients.
BCBSA represents 31 independent insurers, and the analysis attributed $653 million to secondary diagnoses specifically, with $942 million in total added costs versus 2023. The tools in question include ambient scribes that draft notes from patient conversations and AI systems that scan records for missed conditions. The association's leadership pointed to the disconnect between diagnosis rates and treatment patterns as central to their concerns.
This story could affect patients, providers, and insurers in different ways. If insurers respond by tightening reimbursement rules or auditing AI-assisted documentation, doctors may face more administrative friction. Patients could see premiums rise if costs are passed along, or they might benefit if AI genuinely captures previously undocumented conditions that lead to better care. The study's findings may also prompt regulators to examine how AI tools influence billing practices, potentially shaping future healthcare technology policy.