UCL Computer Science’s success in Privacy-Enhancing Technologies’ challenge

The project named STARLIT, also comprising Privitar and Cardiff University, will use PETs to help financial institutions identify suspicious transactions while protecting customers’ personal data.

PETs can be used to facilitate privacy-preserving financial information sharing and analytics, allowing anomalous payments to be identified without compromising the privacy of individuals. 

This initiative has the potential to revolutionise how suspicious financial transactions are identified, but can also be useful in other areas such as a malware detection service which could send a query to a collection of databases held by independent antivirus companies to find out if they consider a certain application as threatening, or by a bank to gather information about certain customers from various partners to improve its risk management of loans.
 
Project participant, Professor Steven Murdoch, said: “PETs show great promise but are not widely used. This competition will demonstrate how effective these technologies are and will help them be used in the financial industry and beyond.”
 
In total, 12 teams were selected as winners in phase one of the PET challenge, receiving a total of £138,000. 

The second phase of the challenges, which began in November, will see participating teams build the solutions discussed in their technical papers, and these insights will be shared with stakeholders, such as government agencies. Participants in the second phase will compete for prizes for a combined total of £803,000.

The prize challenges have been developed as part of a joint effort between the United Kingdom and the United States with the intention of ‘transforming financial crime prevention and boosting pandemic response capabilities through privacy-preserving federated learning.’

The United Nations estimates money laundering costs up to $2 trillion annually, which impacts economic activities and finances organised crime.