London, June 20, 2026, 15:02 BST
- AI is making it easier for fraudsters to draft fake medical files, build fake identities and call insurers in bulk, which is pushing up claim screening costs.
- Gallagher is seeing more policy wordings that don’t directly include or rule out AI-related losses, leaving what some in the market are calling a “silent AI” risk. Gallagher
- Insurance execs say having people involved is still key even as more companies turn to AI for fraud checks, claims work, and protecting themselves from tech outages.
AI is moving past productivity projects at insurers. It’s turning into a claims risk, triggering liability worries and giving criminals on health plans a low-cost way to fake evidence.
AI is making healthcare fraud easier to pull off. Now it takes less skill. A large language model can whip up fake medical records for treatments that didn’t occur. AI bots can hit up an insurer with thousands of calls — no human needed.
Kurt Spear, vice president of financial investigation and provider review at Highmark, said the industry knew AI would be used by fraudsters and “now we’re starting to see that.” The Pittsburgh insurer uses tools to detect AI-driven fraud and is rolling out new tech to catch medical-imaging anomalies to the pixel, according to TribLive. TribLIVE Community
Call-center threats are rising quickly. Pindrop’s tech, used by big health insurers, saw clients get hit with 15,000 bot calls in just a few months, according to Jason Barr, Pindrop’s vice president of healthcare. UPMC and Pennsylvania’s Medicaid agency told TribLive they haven’t seen much AI-driven fraud, showing the risk isn’t spread evenly but isn’t far off.
Healthcare fraud costs the industry tens of billions every year, the National Health Care Anti-Fraud Association says. The Justice Department’s 2025 national healthcare fraud takedown charged 324 people tied to over $14.6 billion in planned losses. Generative AI only adds to the existing risks, according to the .
The liability side is shifting too. Gallagher said this week New Zealand businesses are looking at a new type of AI “silent risk,” similar to what happened with cyber insurance in its early phase, when losses sat in existing policies. Datacom research shows 87% of New Zealand organisations now use some form of AI. That’s up from 66% in 2024 and 48% in 2023. Insurance Business
Gallagher says about 20% of insurance professionals in its latest AI adoption survey reported insureds took economic losses or made claims on AI risks over the past year. Of those, just over half got the full amount covered, 44% were partly covered, and 3% saw no coverage at all. Paige Cheasley, who leads Gallagher’s Canada national technology practice, said, “it can be tricky to attribute losses directly to AI.” Gallagher
The market problem is here. A bad AI result could hit cyber policies, professional indemnity, product liability, employment practices liability, or D&O insurance, which shields company leaders from some suits around management calls. Gallagher’s John Farley said using AI more could lift both the number and size of claims.
Automation isn’t the full answer for international private medical insurance. Algirdas Dineika at WTW told a Health & Protection briefing that insurers still need staff who can work with agentic AI—technology able to handle tasks within workflows. UnitedHealthcare Global’s Janette Hiscock, who leads Europe, Middle East and Africa, was plain: “AI is not infallible, it is hackable.” Health & Protection
But bumping up screening isn’t a silver bullet. More AI use can mean more false positives, slower payouts for legit claims, and opens up privacy or fairness issues. Wide exclusions risk leaving companies with gaps in coverage they thought they had. Joanne Buckle, principal and consulting actuary at Milliman, said AI advances are outpacing what organisations can manage.
AI is boosting the number and detail of fraudulent claims for health insurers, but it’s also one of the only ways to catch those scams fast. Insurers that can link image forensics, voice authentication and claims analytics to proper clinical review and clear audit logs—not just the latest fraud tech—will probably have the edge.
AI is turning up in insurance claims work as a regular tool for regulators. The National Association of Insurance Commissioners put out a Journal of Insurance Regulation paper showing 65% of health insurers are using or planning to use AI or machine learning to spot fraud. That tech is backing up human decisions more than replacing people, the report said. It’s not some distant shift—it’s looking like the next claims expense.
