Codoxo has launched a new artificial intelligence (AI)-powered deepfake detection tool designed to help US health plans identify fraudulent medical documentation before claims are paid.

The solution, called Deepfake Detection, analyses medical records, clinical notes, diagnostic images, and claim context to identify AI-generated or manipulated evidence submitted in support of claims. The company says the tool is already being deployed with health plans across the US as generative AI accelerates the scale and sophistication of healthcare fraud.

Healthcare fraud is already a multibillion-dollar issue, but generative AI is making documentation fraud easier to produce and harder to detect. Fraudsters can now generate convincing progress notes, clinical narratives, and diagnostic images within minutes, creating synthetic records that traditional rules-based systems and manual review processes struggle to identify.

“Fraudsters are adapting faster than legacy defences can respond, and healthcare’s documentation-heavy workflows make payers uniquely vulnerable,” said Dr Musheer Ahmed, Founder and CEO of Codoxo. “Deepfake Detection is designed 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.”

The system flags high-risk cases in seconds and assigns risk indicators to help payment integrity teams and special investigation units prioritise investigations.

“As generative AI becomes more accessible, verifying the authenticity of medical documentation at scale is becoming increasingly complex,” said Kurt Spear, Vice President, Financial Investigation and Provider Review at US health insurer Highmark. “Healthcare organisations need new approaches to identify synthetic or manipulated documentation earlier in the process in order to protect payment integrity and reduce downstream risk.”