Leveraging EGo and Intel SGX to detect insurance fraud

Accenture Labs conducts research on privacy-enhancing technologies, including confidential computing. One of the key use cases to be solved with such technology was the detection of insurance fraud. They conceptualized a platform where insurers could submit claims (e.g., for car accidents) to cross-analyze the data with different insurers and highlight potential fraud. The challenge was to make sure that neither sensitive customer data (e.g., names, addresses, banking details) nor company data was exposed.

Accenture built a prototype of such a platform by leveraging Intel SGX and Edgeless Systems’ open-source software EGo. EGo is a framework to easily develop confidential-computing apps in Go. Accenture developed a platform that securely runs inside SGX enclaves and keeps data completely invisible to any party—even during processing. This security is verifiable through a process called remote attestation, using cryptographic certificates to check the integrity of the software. The platform is designed to cross-examine all data to identify patterns, such as multiple instances of identical key indicators (e.g., phone numbers). If the claim submissions exceeded a pre-established threshold, the platform would alert the user.

Leveraging Intel SGX enclave technology and the EGo framework, Accenture Labs was able to quickly develop a secure platform to detect insurance fraud. This solution has been designed to prioritize data security and privacy and can be integrated by insurance companies worldwide.  

You can watch a webinar on this case study below and you can find other confidential-computing use cases in the financial sector here. If you are interested in a demo, please contact us here.