The Science Of Fighting Fraud

 

A psychologist, biologist, anthropologist and physicist walks into a high tech cybersecurity firm….

Seriously.

Psychologists, biologists, anthropologists and physicists are among the six or seven profiles who fraud technology company, Forter CEO Michael Reitblat tells MPD CEO Karen Webster, make the most effective and expert fraud analysts.

But Forter, admittedly, has a different filter that it uses to vet its fraud analyst candidates.

First, Forter’s fraud decisions are 100 percent made “by machine” and in real time. So, instead of hiring analysts who manually review exceptions or only software engineers who develop data algorithms based on past activities and assumptions, Forter’s fraud analysts must teach the system what they know. And, since each of these scientists are trained to look at data patterns differently, each is reviewing data using a unique lens. Those learnings update Forter’s fraud engine five to seven times a day. By contrast, Forter says that most fraud prevention systems are based on complicated models that are updated once every three to six months.

“We look for [candidates with] extremely high analytical skills,” Reitblat said, which he explained means “looking for lots of details and finding the pattern that connects them. You also have to have really creative thinking … You have to be able to think of different scenarios.”

In a typical test for an analyst, a test that only one out of every 100 might pass, a candidate is given roughly 60 percent of the information needed to crack the case. It’s up to them to connect scenarios, analyze different outcomes and make assumptions to determine if their data corresponds with that outcome.

“It’s important to have a really good eye for details. There’s a lot of little things you can miss, and then they can turn the whole story around. We can’t have fraud analysts that miss things,” Reitblat said.

And while there are many good reasons why that is, there’s one in particular that’s unique to Forter’s business: Forter’s the one on the hook if it approves fraudulent transactions. In the most basic terms, the combination of the observations of these highly specialized humans teach Forter’s computers to understand bad guy behavior and then, Reitbtlat says, recognize fraud faster and stop it in its tracks.

But physicists and fraud? Don’t physicists look at matter and try to understand the characteristics of the particles that comprise it?

Reitblat explains: “If a physicist sees something in the world and they can’t explain it, they then conclude that the theory is wrong, not the observation. And that’s why, he says, physicists make interesting analysts.”

OK, but what about anthropologists — those people that study how and when human beings evolved to walk upright?

Forter says that it all goes back to identifying problems and patterns and having a formula that explains the entire process. That means, in theory, that a person with the ability to understand behavior across multiple countries and cultures may have the distinct skill set for how to break down patterns, understand how behaviors differ by region and see how those behaviors could influence how a hacker would think from one country to another.

“Our approach is to go as broad as possible and create a really, really large assembly of tight nets that, through combination of them, will just be way more accurate using a variety of disciplines,” Reitblat explained.

Analyzing The Science Of How Fraud Takes Root And Goes Viral

And then there’s the sociological and biological side of analyzing fraud, which is all about determining how fraudsters and their bad guy behavior have evolved over time, including their natural and learned behaviors. After all, as Reitblat explained in an earlier interview with PYMNTS, today’s highly specialized hackers are training those that will dominate the field tomorrow.

So, studying the evolution of that behavior as it evolves is key to understanding how the patterns related to hacks and security breaches develop, Reitblat believes.

“[Fraud] is a distributed industry made up of many different fraudster profiles,” he emphasized. “So, having one, and only one, type of fraud analyst won’t stop all the many different types of fraudster and, more importantly, not allow us to recognize how their patterns change and develop over time.”

Leading Reitblat and Forter to eschew any “generic description of a fraud analyst” and instead create a “surgically assembled team.”

It also doesn’t hurt, Reitblat says, that a lot of Forter’s founding team was based in Israel, a country where compulsory military service gave those physicists and anthropologists the training and attention to detail needed to recognize the patterns for cyberfraud.

Reitblat describes what’s taught in Israeli military intelligence training is how to recognize patterns related to terrorism, weapons smuggling and cybercrime. Behavior that Reitblat said is similar, in concept, to tracking fraudsters’ activity across an ecosystem of bad actors intent on doing harm.

“We try a lot of things that don’t work to find out those things that do,” Reitblat said.

Beyond Fraud: The Bigger Picture

Cybersecurity — both from a data and fraud perspective, as well as from a national security standpoint — has become practically a daily buzzword in nearly every major economy. That’s made countries become increasingly heightened to cyberthreats and ensuring systems are secure and kept out of the wrong hands.

“The world is becoming more and more digital, and it’s not only in commerce,” Reitblat said. “Through the Internet, we can probably control 90 percent of our functions — from water to electricity to national security to anything you can think of. Securing them is crucial. The price [in attacks] is not just for money, but it’s also for difficult control of systems.”

But how each nation approaches the subject is still drastically different. For military-grade technologies (Israeli military or NSA), those organizations work in similar matters, Reitblat told Webster, but across the fraud industry itself, there are many differences.

That’s where Reitblat would like to see changes most, and it could perhaps be a region Forter can have more of a presence in.

“In the U.S., there isn’t enough of a migration of people from defense- or intelligence-related professions into the civilian life. I would like to see more ex-NSA agents in security startups or, at least, cyberunit experts or just scholars that study those things in universities,” Reitblat said.

“I would like to see it get more integrated into Silicon Valley … Every problem probably can’t be solved with engineers. Diversity is good for you, not bad for you,” he concluded.

Perhaps someday, we’ll see psychologists, biologists, anthropologists and physicists calling that region home.

Something everyone but the bad guys would be very happy to have happen.