Going Vintage at MRC 2017
Latest posts by Paul Bookstaber (see all)
- A New Timescale for Fraud Science: Insights from Our CSO - October 17, 2017
- Operationalizing Machine Learning for Fraud: Human + Machine Learning is the Best of Both - October 6, 2017
- Easing Into Machine Learning for PSD2: An Interview with Our VP of French Sales - September 13, 2017
In a future-focused industry where features and concerns seem to appear from somewhere beyond the horizon of the present moment, it was a surprise that the Merchant Risk Council main event this year focused on the past. Paper shredders, anyone? Check fraud? Copy machines?
Maybe that’s all fitting. Everything old is new again, as they say. And as “Art of the Steal” author Frank Abagnale said this morning in his keynote at this week’s annual trade show in Las Vegas, there’s one thing that engineers and technology will never solve: social engineering.
He talked about the types of shredders companies should buy, about a certain pen whose ink couldn’t get removed by a check forger armed with acetone, about how copy machines get resold with their hard drives still containing stored data.
Companies like Feedzai are pursuing a future-proof system, one that’s so agile it can handle use cases and channels that might not even exist yet. But the wall of human creativity — particularly the creativity of criminals — might be the last frontier, and AI may never surmount it.
Mike Kereseman, founder and former CEO of Cardinal Commerce, introduced Mr. Abagnale with a history lesson. He traced our ten thousand year journey from bartering with barley, to the origins of a trade exchange in the middle ages, to what finally became our modern system in 1971, with fiat money.
Fiat money isn’t back by anything tangible. Instead, it’s backed by a perception of public confidence, the expectation that the government behind it will continue to prosper into the future. So fiat money is backed by faith and trust. That makes the core mission of a merchant to support a world of faith and trust, to create a solid ground where buyers can buy and sellers and sell.
Mr. Abagnale picked up the thread of principle and spooled it into a kind of secular sermon. “We live in an extremely unethical society.” You might remember Mr. Abagnale as the true story behind Steven Spielberg’s “Catch Me If You Can.” He’s changed his ways. “We will never make a dent in white collar crime,” he said, “Until we bring character and ethics back into the home, the workplace, the company.”
Wait a minute. I came to MRC to talk about algorithms, use cases, and multi-tenancy. So why were we talking about faith and trust, character and ethics? Well, if every breach happens because humans are doing something they shouldn’t, or not doing something they should, then detecting fraud is about detecting humans. And while Moore’s law might have the rate of change doubling and doubling, human nature (you could argue) is constant.
AI is changing, growing, and becoming more powerful. But its subject, humanity, remains the same. So I thought the morning was an interesting reminder that humans don’t serve the machine. The machine serves the human. The question then becomes: what kind of human is being served by this growing power of machine learning? In a world where fraudsters are just as smart, creative and motivated as the good guys, machine learning is only the latest entry in that eternal arms race between the good guys and the bad guys. And to piggyback on an old Cherokee legend, in the battle between the evil wolf and the good wolf, the wolf that wins is the one we feed.