Press Releases

The Start-Up That Reveals the Hackers of Today’s Accounts

March 17, 2020

“We reveal the immense and silent business of money laundering”

“Everyone speaks about artificial intelligence… but making computers capable of human intuition and common sense was considered black magic and simply impossible!” Mark Gazit may sound like he’s hyperbolizing, but the CEO of ThetaRay, an Israeli start-up based in Hod HaSharon, New York (JVP Cyber Center recently inaugurated in Manhattan), London, Mexico City and Singapore has no better way to summarize the extraordinary achievement of the Yale mathematician Ronald Coifman and Tel Aviv University Professor Amir Averbuch. These two professors have developed a series of algorithms capable of revealing the unknowns from very large and complex data sets. “Artificial intelligence,” explains Gazit from the New York co-working space, “can only detect what it has been taught to know. Once it detects something different, it can’t differentiate, since it doesn’t know what it represents. Because the system isn’t aware that it is receiving new information, that’s where crucial data falls between the cracks.” This most certainly is not an easy puzzle to solve, especially when dealing with real life situations.

Initially, ThetaRay operated in the fields of cyber security and critical infrastructures, specifically in nuclear plants and airlines. “A few years ago we stumbled across the world of financial institutions,” says the CEO, “and we noticed how much the crime rate in this sector has developed and changed within the past five years”.

Traditional armed robberies, as we are use to seeing in Hollywood films, no longer occur as we nowadays face a wave of “new thieves” such as hackers, who are able to gain access to confidential information and can easily engage in criminal activity all from behind a computer screen. Not only do todays criminals take advantage of money laundering for personal use, they now resort to severe offenses such as human trafficking, drug trafficking, sexual exploitation and the financing of terrorism.

All that a “modern day criminal” needs in order to take advantage and hack into a system, is a simple server installed in a remote country somewhere around the world. That way, the criminal can gradually steal a few cents from hundreds of thousands of different accounts, which eventually leads to millions of dollars stolen in theft. That’s how millions of dollars manage to disappear without being able to locate the culprits. “The existing systems,” continues Gazit, “used to battle money laundering no longer work since they are still using “rule based” systems to target the behavior of the “bad guys” as used in the traditional banking world. The attacks we are facing now are completely new, and there has been no way to teach basic AI systems to recognize this odd behavior, until now…”

Based on deep machine learning and algorithms developed over the course of 15 years, ThetaRay’s system can detect these “new behaviors”. The existing systems in banks monitor money strictly when it enters or leaves an account, but we now know that a network of accounts in different countries can also be involved in money laundering. ThetaRay’s intuitive AI solution can detect “blind spots” (also know as “unknown-unknowns”) in large and complex sets of big data by detecting anomalies within the data. In essence, and as portrayed in the 2002 film directed by Steven Spielberg, “Minority Report,” ThetaRay’s system can detect financial crime before it occurs. As an example, ThetaRay was able to identify a network of tens, if not hundreds of thousands of people, who laundered money to finance ISIS. Each transaction appeared to be innocent as a small amount of merely 10 or 15 dollars, 20 euros or 5 Swiss francs was transferred each time, but the ThetaRay system was able to identify the suspicious pattern. “As far as I know,” concludes Gazit, “we are the only ones in the world able to detect a crime like this.”

This page has been translated from its original article: 

was shared 0 times