Hospitals and insurers deploy AI to fight over bills—patients lose $2.3B

How the AI billing war inflates diagnoses and denies care

After a patient is discharged, hospitals generate a superbill that gets translated into diagnosis codes. Those codes determine what insurers pay. Hospitals can increase reimbursement by identifying secondary diagnoses—complications or comorbidities documented in the medical record that justify billing patients as sicker.

This coding was once done by humans. Now hospitals deploy AI-powered ambient listening software that records physician-patient conversations and automatically populates the electronic health record with diagnoses. These systems, marketed to reduce physician burnout, have a side effect: they identify and document secondary diagnoses more aggressively than clinicians would note manually.

The data are stark. Blue Cross Blue Shield analyzed tens of thousands of maternity admissions nationwide and found that at hospitals with the fastest adoption of AI coding tools, the proportion of patients coded with acute posthemorrhagic anemia (serious post-delivery bleeding) jumped from roughly 4% to more than 12% between 2022 and early 2025. At hospitals without those tools, the rate barely moved (per BCBS analysis).

Yet the increase in diagnoses was not matched by increased treatment. Blood transfusion rates, which you would expect if more women were truly hemorrhaging severely, remained flat. When auditors examined records at one hospital, fewer than 20% of the coded cases met clinical criteria for the diagnosis. BCBS estimated AI-enabled coding added $22 million to maternity costs alone at the studied hospitals in a single year, with nationwide effects potentially reaching $2.3 billion (company-reported).

Massachusetts offers a second proof point. Hospitalizations for septicemia have more than tripled since 2010, reaching more than 42,000 cases in the year ending last September. Sepsis is now the third-leading cause of hospitalization in the state. But researchers, insurers, and even some hospital clinicians are skeptical that Massachusetts has actually become that much sicker. The Massachusetts Health Policy Commission documented that the highest-severity cases—the ones that fetch the largest reimbursements—are growing fastest, without corresponding increases in length of stay or intensive care utilization. Sepsis, it turns out, is one of the most lucrative secondary diagnoses in the Medicare billing system.

Meanwhile, insurers have deployed their own AI systems—program integrity platforms that detect patterns suggesting upcoding, flag suspicious claims, and deny reimbursements automatically. Both sides call their work clinical and responsible. Both sides are training algorithms to fight over the same patient record.

The friction falls on patients, not the institutions fighting

Health systems collectively spend more than $140 billion annually on revenue cycle management alone. Administrative expenditures consume more than 40% of total hospital spending. Hospitals now conduct nearly half a million post-claim inpatient reviews per year for a single large insurer, each one a small battle in a war neither side can win decisively because the other side adapts and learns.

The patient—the person who gave birth in a car, or had sepsis, or spent 56 days in a neonatal intensive care unit—is not at the table. Consider Bisi Bennett, who delivered her son prematurely in her husband's car and received a $550,000 hospital bill with a proposed payment plan of $45,843 per month. It took a journalist's intervention and human review to cut the bill in half. The hospital's system had swallowed the electronic medical records and automated, algorithmically driven computers spit out numbers that took on a life of their own. Those numbers crashed against the insurer's own systems, which fired their own numbers back.

This is not a single billing error. It is an arms race in which every step generates friction: a test requires up-front payment, a referral hits an insurance wall, a prescription is on the wrong formulary tier, a claim gets denied and appealed, a bill arrives, then another, then a collection notice. Each individual obstacle is manageable in isolation. But they compound. Patients skip follow-up visits. They don't fill prescriptions. They decide quietly that the system is not worth fighting. When they reappear, their conditions have often progressed. The financial side effects of care have become clinical ones.

A different architecture is possible—but it requires ceasefire thinking

One proposal currently being explored by some organizations would fuse the two competing AI systems into a single adjudication engine. It would ingest clinical records and billing data simultaneously at the point of discharge and issue a single consensus payment decision in near real time. Rather than having hospital AI and insurer AI fight over the same record for months, a unified system would evaluate both the provider's right to accurate reimbursement and the payer's obligation to appropriate payment at the same moment.

Most claims would pass through immediately, giving hospitals faster and more predictable cash flow. Claims with genuine discrepancies would be flagged for immediate human review, with documentation already in hand. This is not a panacea. It requires unprecedented data-sharing between institutions that have spent decades treating each other as adversaries. It demands governance frameworks that don't yet exist. And it leaves open harder questions about underlying price levels that will require legislative action, not just technological creativity. But it points toward something important: the possibility of a system that settles the question of what care costs before the patient ever sees a bill, rather than after months of algorithmic trench warfare.