UnitedHealth faces lawsuit over AI model with 90% error rate
Summary
UnitedHealth is facing a lawsuit alleging that the company has been using an AI model with a 90% error rate to deny patient care. The lawsuit is currently under the oversight of federal courts as part of a class action that challenges the insurer's use of AI in making coverage determinations for Medicare Advantage plans. In response, UnitedHealth asserts that its AI tools are intended to support, not replace, the clinical decisions made by physicians during the claims process.
Analysis
Ars Technica: Ars Technica is an online technology news publication focused on science, policy, and tech developments. It reported on allegations in a lawsuit claiming UnitedHealth deployed a flawed AI system to deny care to Medicare Advantage beneficiaries. UnitedHealth: UnitedHealth Group is a major US health insurance provider offering Medicare Advantage plans and related services through its UnitedHealthcare division. The company faces ongoing class-action litigation alleging that it used an AI model to determine coverage for post-acute care instead of relying on physicians as stated in policy documents. Recent court rulings in 2026 have compelled production of documents concerning the model's development, training, and application in claims decisions. Lawsuit Status: Federal courts continue to oversee discovery in the class action challenging UnitedHealth's use of AI for coverage determinations in Medicare Advantage plans. AI in Healthcare: UnitedHealth maintains that AI tools support rather than replace clinical decisions made by physicians in claims processes.
Categories
aitech
Related sources
- https://law.justia.com/cases/federal/district-courts/minnesota/mndce/0:2023cv03514/211721/162/
- https://arstechnica.com/science/2023/12/humana-also-using-ai-tool-with-90-error-rate-to-deny-care-lawsuit-claims/
- https://endpoints.news/unitedhealth-group-lays-out-progress-in-ai-push/
- https://www.statnews.com/2025/05/19/unitedhealth-algorithm-ai-denials-of-care/
- https://www.constellationr.com/insights/news/how-unitedhealth-group-sets-its-ai-stack-targets-1000-use-cases
- https://www.fool.com/investing/2026/07/14/unitedhealth-is-investing-billion-in-ai-and-target/
- https://www.physicianspractice.com/view/ai-s-hidden-hand-decoding-prepayment-denials-and-the-evasion-of-high-dollar-claims
- https://www.mashupmd.com/unitedhealth-uses-ai-model-with-90-error-rate-to-deny-care-lawsuit-alleges-3/
- https://www.afslaw.com/perspectives/alerts/federal-court-orders-broad-discovery-against-uhc-ai-coverage-denial-lawsuit
- https://www.courthousenews.com/federal-judge-dismisses-several-claims-in-ai-denial-lawsuit-against-uhg-but-case-will-proceed/
- https://www.forbes.com/sites/daraabasiita/2026/06/09/the-algorithm-that-counted-on-no-one-appealing/
- https://www.fiercehealthcare.com/payers/rural-health-system-ballad-health-sues-unitedhealth-over-medicare-advantage-claims-denials
- https://data.larridin.com/data/ai-tracker/company/unitedhealth
- https://bestofai.com/article/unitedhealth-uses-ai-model-with-90-error-rate-to-deny-care-lawsuit-alleges
- https://ghostarchive.org/archive/7gsC2
- https://makerchecker.ai/insights/unitedhealth-nhpredict-ai-medicare-denials/
- http://www.modernhealthcare.com/insurance/mh-unitedhealth-ai-optum-sandeep-dadlani/
- https://www.tresslerllp.com/thought-leadership/estate-of-gene-b-lokken-et-al-v-unitedhealth-group-inc-ai-risks-in-medical-insurance-coverage-disputes/
- https://www.classaction.org/news/unitedhealth-groups-flawed-ai-model-wrongfully-denies-elderly-patients-post-acute-care-coverage-class-action-alleges