The AI solution to know your customer and anti-money laundering checks. (Illustration : Maison Moderne)

The AI solution to know your customer and anti-money laundering checks. (Illustration : Maison Moderne)

Advanced computer technology can greatly facilitate the financial sector’s obligation to know whom they are doing business with. Cutting-edge approaches, including artificial intelligence (AI) and other sophisticated tools, can be put to work on helping firms with their anti-money laundering requirements.

A team at the University of Luxembourg’s (SnT) has developed a know your customer (KYC) and anti-money laundering (AML) search engine platform called . Their solution is powered by their unique technology that leverages the “wealth of data that is publicly available, including watch lists, business registers, social media and much more, in order to produce a unified digital footprint,” according to , the CEO of digitalUs.

Georgios Varisteas, Beltran Borja Fiz Pontiveros, Raphaël Frank and Oxana Turtureanu. (Photo : SnT)

Georgios Varisteas, Beltran Borja Fiz Pontiveros, Raphaël Frank and Oxana Turtureanu. (Photo : SnT)

digitalUs will soon finish the proof-of-concept stage of the . With the assistance of SnT’s Technology Transfer Office, they aim to be the next spin-off from SnT.

Know your customer

Aware of public concerns about the misuse of such information, the team identified the financial sector’s KYC and AML regulatory requirements and developed a tool that could be commercialised to share their technology. Financial companies have a legal obligation to perform due diligence on new customers,” said , a professor at SnT’s (SEDAN) research group and co-founder of digitalUs. 

The researchers felt they could improve on existing services targeting KYC and AML. “With AI technology, we get a much better result because we are using text and image analysis,” said Frank. He explained that someone might use different pictures of themselves on various social media platforms or have a very popular name. With the ability to recognise faces or use natural language processing, which allows for keyword analysis, digitalUs can “Pinpoint the exact John Smith we’re talking about,” he added.

Entity resolution

Frank gave the example of someone who perhaps has a LinkedIn page with several companies mentioned, but no additional information on them is provided. “In other repositories we may find additional information from these companies, including beneficial owners, which allows us to connect the same individual across multiple data sources,” he explained. “This is called entity resolution and is the technology we want to showcase.” This technology can provide a solution for the time-consuming manual tasks that financial institutions are obligated to perform to prevent money laundering and fraudulent activity. The tool can automate the KYC process, which would significantly reduce the time it takes to screen each individual, and be able to find out swiftly if they have any connections or activity that may signal a red flag.

Teach them how

One key distinction between digitalUs and competing players is their approach to storing information. Fiz explained that most companies create a ‘data silo’ full of third-party information to supply an answer when searched. “We are going to teach our platform how to search the data online and put the profile together without needing to store all that information,” said Fiz.

digitalUs currently has three full-time staff working on its launch, including fellow co-founders , the Chief Product Officer, and , the Chief Technical Officer. Speaking about the next phase of the project, Fiz said, “We are very close to the end of the proof-of-concept, and we are already starting to prepare for the launch of this spin-off company now. The feedback received by our commercial supporters during development was crucial to find product-market fit.”

Time to go to market

Fiz shared how proud the digitalUs team are of their product. “We have one of those tools that we hope will appeal to the client immediately. They should see the added value and say, ‘I want this!’”.

“We have already signed some early adopters and will continue over the next few months to acquire clients and seek out investment to achieve higher growth rates,” said Fiz, who concedes digitalUs will have to prove itself. “The aim is to be established by the end of the year, at least in the Luxembourg ecosystem.” He also believes the tool has the potential to one day help citizens manage their own digital footprints and help raise awareness of the data they share online.