US President Joe Biden could soon be paying homage to novel fraud busting technology from Cambridge innovator Featurespace.
The company is in the running for glory in the transatlantic privacy enhancing technologies (PETS) challenges jointly steered by, among others, Innovate UK and the White House Office of
Science and Technology.
Featurespace’s nextgen AI technology is set to be showcased at the second Summit for Democracy, to be convened by President Joe Biden in March. Contenders are competing for
cash prizes from a combined UK-US prize pool of $1.6 million.
The challenges are focused on accelerating the adoption and development of an emerging group of data-driven technologies known as PETs. The technologies have the potential to unlock
transformative insights from valuable datasets to tackle global societal challenges, while at the same time preserving citizens’ privacy.
Featurespace is a front runner in the financial crime prevention category. Having kept clients and their customers alert to scams during the pandemic, Featurespace is now focusing on the ‘scamdemic’ – a virulent global outbreak of fraudulent skulduggery.
The aim of the financial crime prevention category is to develop solutions able to train machine learning models to detect anomalous transactions whilst preserving the privacy of individuals’ financial information contained within the datasets.
For nearly a decade, Featurespace has been at the forefront of modernising financial crime prevention in financial services institutions (FIs) around the world. Its next generation machine learning models provide industry leading predictive performance. Privacy preserving technologies build on this by allowing collaborative learning, which Featurespace believes will be one of the vital leaps forward in the fight against financial crime.
The approach uses deep learning, the successful utilisation of which has in recent years enabled Featurespace to push the limits of what detection systems can achieve using existing data
sources. Privacy preserving techniques, such as de-identification, local differential privacy, resampling, and k-anonymity are baked into the design for both training and inference stages.
This means that all communications between nodes are privacy preserving and the privacy of citizens is never compromised.
Featurespace is prototyping its specific solution in line with the PETS challenge guidelines at the moment and will submit the finished solution in advance of the official deadline on January
24. The enhanced anti-scam solution cannot come soon enough as the United Nations estimates that money laundering costs up to $2 trillion each year.
Near the back end of 2022, Featurespace secured unspecified PETS funding from the UK and US governments to build a new type of AI system to help banks and payments service providers detect financial crime whilst protecting data privacy. The funding has enabled Featurespace to develop privacy preserving solutions that allow AI models to be trained on sensitive private data – important for revealing criminal activity –without organisations having to reveal, share, or combine their raw data.
The company has applied cutting-edge federated deep learning incorporating privacy enhancing techniques such as k-anonymity and local differential privacy to tackle financial crime. On top of the UN figures, trade body UK Finance predicts an ‘epidemic of fraud’ due to increases in scams, which totalled £580 million lost at the last count in 2021 – representing a 40 per cent year-on-year increase in this type of crime.
Dr David Sutton, Featurespace’s Director of Innovation, says: “UK and US governments want banks to work together to stop fraud and money laundering. “This type of privacy preserving collaborative AI is a hard problem that no-one has yet solved. We are confident we have met this challenge. We’re the only company in this project that has deployed innovative technology to fight worldwide financial crime – and we have the banking customers to prove it.
“A successful outcome of this project is to make money laundering across borders and between banks much more difficult. If you make it harder to launder money you make criminal activities less profitable. This will benefit businesses, society and consumers.”
Featurespace’s solution is intended to help PSPs and banks tackle scams, including cross border money laundering. The company invented Adaptive Behavioral Analytics and Automated Deep Behavioral Networks and is the first to profile both genuine and fraudulent behaviour to identify and block criminal activity in real time.
More than 70 direct customers and 200,000 institutions have put their trust in Featurespace’s technology including HSBC, NatWest, TSYS, Worldpay, Marqeta, Contis, Danske Bank, Akbank, Edenred and Permanent TSB.
Featurespace has a team of more than 400, operating globally from seven locations.
Read the full article >>