Insurance Analytics

Insurance fraud detection driven by image recognition​


Insurance fraud is an evergreen issue for the insurance industry. It constantly evolves as the types of fraudster behavior become more diverse and sophisticated. 

According to the latest research, undetected insurance fraud accounts for up to 10% of all claims expenditure in Europe. With the introduction of automated (STP) claim processing, fraud attempts are expected to rise even more.


By harnessing the power of AI & Image Recognition, we are able to automatically detect duplicate and exaggerated claims, without the need to share any private data.

In a fully GDPR-compliant process, each submitted image is converted into a unique, 100% depersonalized fingerprint. Those fingerprints are then sent to a common and fully anonymized market database where they are compared with each other. During this process, pairs with unusual similarity are identified with very high accuracy and a red flag is raised.

This approach allows us to completely eradicate duplicate and exaggerated claims from any given insurance market.


  • Total direct savings for uncovered frauds in millions of EUR.
  • Fully automated and GDPR-compliant process, which enables comparison of images without sharing any personal and sensitive information​.
  • Quicker and more efficient claims handling and investigation processes​.


Contact us

Tervel Šopov

Tervel Šopov


Tervel is a consulting director responsible for AI&Data Strategy, Data Science and Machine learning market offerings in Deloitte Central Europe. Tervel has worked in the space of analytics, data and A... More

Jim Vazquez

Jim Vazquez


Jim has comprehensive experience in Data Analytics, Information Technology, and Forensic Technology. His experience in international projects spans 5 continents. As Director of Analytics and Forensic ... More