Health Plans Deploy AI to Catch Synthetic Medical Records in Claims Fraud
US health insurers are adopting AI tools to detect forged medical documentation as generative AI makes healthcare fraud easier to execute at scale. Codoxo launched its Deepfake Detection system to identify manipulated records and AI-generated diagnostic evidence before claims are paid.
The tool analyzes medical records, clinical notes, diagnostic images, and claim context to flag synthetic or altered documentation. It assigns risk indicators within seconds, helping payment integrity teams and special investigation units prioritize cases.
Healthcare fraud already costs the system billions annually. Generative AI has accelerated the problem - fraudsters can now generate convincing progress notes, clinical narratives, and diagnostic images in minutes, creating records that traditional rules-based systems miss.
Dr Musheer Ahmed, Codoxo's founder and CEO, said the company designed the tool to help payers fight AI-assisted fraud with AI. "By identifying synthetic or manipulated medical documentation earlier, we can strengthen payment accuracy, reduce downstream recovery costs, and protect provider relationships," he said.
Kurt Spear, Vice President of Financial Investigation and Provider Review at Highmark, a major US health insurer, said healthcare organizations need new approaches to verify documentation authenticity at scale. "As generative AI becomes more accessible, verifying the authenticity of medical documentation at scale is becoming increasingly complex," he said.
The system is already deployed across US health plans. The technology addresses a specific vulnerability: healthcare's documentation-heavy workflows make insurers targets for fraudsters using generative AI and LLMs to create convincing false records.
Healthcare organizations interested in AI for healthcare applications should understand both the risks and detection methods that new tools employ.
Your membership also unlocks:
