Perhaps you will agree, the most head-scratching, migraine-inducing aspect of Sanctions Screening is free text fields, like SWIFT: 50A, F or K and SWIFT: 59 or 59A. Complidata's NER (Named Entity Recognition) tool utilizes deep learning techniques to extract and resolve significant entities, including names, companies, and locations, enabling comprehensive understanding and analysis. Just enter in the unstructured data... 

...and press "Submit"! 

Great events in New York!

Despite the snowstorm, IIBLP, LIBF, Coastline and S&P Global still pulled through and gave us two days of the most stimulating discussions, insightful panels and engaging events. Here are some of the (many) interesting insights and observations our team have gathered:

At S&P Global seminar, where our CEO Ben Arber was a speaker:
💡For DUGs and export controls, identification of shell / front companies remains difficult, yet critical
💡In DUGs screening, goods descriptions remains a big challenge. Even if OCR can achieve great things, vague good descriptions still need human decision
💡The Russian invasion of Crimea in 2014 was the game-changer: unlike in the Iran and North Korea cases, Russia is massively integrated in the global financial system.

At ATFCP Compliance Annual Meeting, where technology and the role it has to play was the overarching theme of many sessions. From the panels hosted by Ben and Alex, as well as the discussions we had with fellow attendees:
🤖 We will increasingly see the use of AI in trade as it increases efficiency. This
doesn’t necessarily means we have to rip and replace - it can be built up from simple OCR
🤖 One of the main obstacles hindering the adoption of AI technology is the concern around data privacy of such tool, a.k.a data leakage.
🤖 Different banks are at different stages of maturity both in their compliance programs and adoption of technology; more needs to be done to help smaller financial institutions with lower budgets

What do you think? Let us know below ⬇️