In order to trace financial crime more effectively, it is essential that organisations be able to share information and data with one another. At the same time, the privacy of ordinary citizens or businesses must not be violated. How can an organisation combat fraud and money laundering without violating privacy? Via privacy by design.
building up anonymised databases
At TNO, we work on different forms of privacy by design. Multi-Party Computation (MPC), is one of these: a smart way to jointly analyse data without having to reveal them. Cryptographic techniques ensure that several parties can analyse data together and draw conclusions, all without ever being able to see each other’s data. With MPC, absolutely no data is revealed, only conclusions based on that data.
Whitepaper: 'Finally, a privacy-friendly way to harness data'.
Discover how to harness data for fraud detection while safeguarding privacy.
Fraud detection must be improved
Fraud has a large impact on Dutch society. It has a disruptive effect and costs the government, businesses and citizens a lot of money. Furthermore, fraud often has an impact on the more vulnerable members of society because money does not end up where it should. During the COVID-19 pandemic, we have also seen the misuse of government subsidies for companies. In order to detect fraud more effectively, more information must be exchanged between both companies and governmental parties.
Fraud detection without infringing privacy
On the other hand, recent court rulings have shown that combining information can quickly lead to a breach of privacy for citizens. MPC offers possibilities to gain very specific insights from the data of these parties without exchanging sensitive data. Moreover, this technique ensures that only analyses agreed upon beforehand can be carried out. This prevents the improper use of personal data.
Want to know more about fraud detection with MPC?
- Discover how you can use data in a privacy-friendly way.
- Find out how MPC can improve fraud detection by securely linking the data.