In brief
Aite Group analyst David Weiss argues that artificial intelligence will not replace bank compliance departments. It will work alongside them. He calls this approach 'bionic' rather than 'replacement'. Current technology is not reliable enough, legal rules are too ambiguous, and tolerance for mistakes is very low. Where AI does help is speed and scale, especially in cutting false alarms and scanning chat and email.
The central tension is between a flood of regulation and alerts that humans cannot keep up with, and the near-zero error tolerance that stops machines from taking over. Weiss expects fewer compliance staff and more automated, drag-and-drop tools. He reads IBM's Promontory deal, Credit Suisse's Palantir venture and NASDAQ's Digital Reasoning partnership as signs this is already happening.
Questions
Will AI replace bank compliance departments?
No, according to David Weiss of Aite Group. He expects a 'bionic' mix where machines handle volume and humans supervise, because the technology isn't ready and the legal stakes are too high.
What is a false positive in bank compliance, and why does it matter?
It is an alert that flags a legitimate transaction or message as suspicious. Weiss says they are a huge problem, and even wire transfers under $10,000 get held up for review.
What can AI do that human compliance staff can't?
Weiss says it is speed and scale: people could do the work given enough time, but they can't keep up with the volume. A machine can learn from past cases and then work through far more of them.
Why can't AI just read all the new financial regulations and turn them into rules?
Weiss says rules come in layers from Congress, regulators like the SEC and CFTC, and self-regulators like FINRA and NFA, and they interpret things differently. Lawyers would still issue opinions on the edges, and mistakes carry a high legal cost.
What does the IBM and Promontory deal signal?
IBM agreed to buy Promontory, a regulatory consulting firm with about 600 consultants who will help train Watson on compliance. Weiss sees it, along with Credit Suisse with Palantir and NASDAQ with Digital Reasoning, as proof that AI in compliance is already happening.
Facts
Stated by the speakers. Each quote matches the transcript, which shows what was said, not that it is true.
IBM agreed to buy the regulatory consulting firm Promontory Financial Group, which has about 600 consultants who are to help train IBM's Watson for compliance work.
“IBM agreed to buy this very large regulatory consulting firm called Promontory that has 600 consultants in its group”
0:00:04 · Penny Crosman · IBM acquisition · 2016
David Weiss of Aite Group has written several research reports on artificial intelligence, which he calls 'inorganic intelligence'.
“David has been doing a lot of research and he's done several reports on artificial intelligence, or as he calls it, inorganic intelligence.”
0:00:04 · Penny Crosman · AI research
David Weiss argues that compliance will use a 'bionic' model of humans working with machines rather than machines replacing humans.
“I think the solution is bionic versus replacement, and I think that'll be the case for some time.”
0:01:00 · David Weiss · Bionic compliance
David Weiss says NASDAQ partnered with Digital Reasoning, a move he links to the need to reduce false positives in surveillance.
“So that's why NASDAQ got together with digital reasoning.”
0:04:00 · David Weiss · NASDAQ partnership
David Weiss says Credit Suisse and Palantir are working on near-real-time anomaly detection in communications.
“And that's certainly one of the things that Credit Suisse and Palantir have in mind with their efforts.”
0:04:57 · David Weiss · Credit Suisse Palantir
David Weiss says the London Whale trading loss was a failure of risk management rather than a failure of compliance technology.
“in the case of the London whale I mean that was just poor risk management and it seems not even paying attention to the traders who were saying enough”
0:05:14 · David Weiss · London Whale
David Weiss says banks have been collecting chat logs from traders for roughly a decade.
“the truth is that data has always been at these firms for a decade. Chat logs have been shipped for a long time.”
0:05:50 · David Weiss · Chat log retention
David Weiss says chat logs are typically sent to the firm at the end of each day, where they may be parsed or simply stored as a text file.
“the logs are shipped at the end of the day, the firm gets it, and they may parse it or they may just persist the text file.”
0:06:34 · David Weiss · Log handling
Penny Crosman says the Dodd-Frank Act ran to more than 3,000 pages.
“Like I think Dodd-Frank was more than 3 ,000 pages. Have the software read all of that”
0:07:20 · Penny Crosman · Dodd-Frank
David Weiss says the U.S. is at the leading edge of implementing the 2009 G20 agreement, with several regulatory bodies on different timetables.
“where we are on the leading edge of 2009 G20 agreement implementation, you've got multiple bodies with multiple time frames.”
0:08:26 · David Weiss · G20 reforms · 2009
David Weiss says the SEC is well behind the CFTC in many areas of swaps regulation.
“the SEC is so behind the CFTC in many areas when it comes to swaps, for example.”
0:08:26 · David Weiss · Swaps regulation
David Weiss estimates that a failure rate of about 1 percent might be acceptable in business-to-business technology, versus about 10 percent for consumers.
“In business to business, maybe it's on the order of 1 % failure rate is acceptable.”
0:10:01 · David Weiss · Failure tolerance
David Weiss cites SEC Commissioner Kara Stein as asking how an algorithm could be certified to meet the fiduciary standard.
“as SEC Commissioner Stein said, how do you certify an algorithm to the fiduciary standard?”
0:10:01 · David Weiss · Fiduciary standard
David Weiss says wire transfers of less than $10,000 can be flagged for anti-money laundering review.
“something gets flagged for review and it could even be something less than $10 ,000”
0:12:08 · David Weiss · Anti-money laundering
David Weiss says machine learning is heavily used in retail banking to detect structuring, which means splitting transactions to avoid reporting.
“when you're looking for structuring, and this is in the retail world, machine learning is huge, as is the flip side of data mining.”
0:12:08 · David Weiss · Structuring detection
David Weiss says AI's main advantage over humans in compliance is speed and scale, not a new kind of ability.
“It's speed and scale. That's the best way to put it. It's not that we can't do it. It's we can't keep up with it.”
0:13:17 · David Weiss · AI advantage
David Weiss describes machine learning as able to learn from flagged cases, predict false positives, and eventually flag likely false positives itself.
“a machine can start to predict when something might be a false positive and then taking another iteration out, the machine can start flagging for itself what it thinks is a false positive versus real”
0:13:45 · David Weiss · Machine learning
David Weiss says the Wells Fargo scandal was not a machine learning or software problem, because people at the bank knew what was happening.
“you can see what's going on with Wells Fargo and it's like you know people knew what was going on there”
0:15:36 · David Weiss · Wells Fargo · 2016
David Weiss says capital markets firms cut headcount by 20 to 30 percent over about two years, mostly in the front office.
“there's a 20 to 30 % headcount reduction in the last maybe two years, and it's mostly been in the front office.”
0:15:57 · David Weiss · Headcount
David Weiss says compliance and risk staff now formally sit on trading desks in seats vacated by traders.
“a number of seats on the trading desk have been taken up by compliance and risk management pros who now formally sit in the front office”
0:16:24 · David Weiss · Trading desks
David Weiss says hiring in compliance has plateaued and he expects reductions through attrition without refilling roles.
“And in the case of compliance, that hiring has kind of plateaued off, and maybe we're going to see some reduction or some attrition and just not refilling.”
0:16:24 · David Weiss · Compliance hiring
David Weiss says Credit Suisse hired Palantir two or three years earlier and later formed a joint-venture startup with it.
“If they could have the foresight two or three years ago to retain Palantir, and then down the road say, this is pretty good, let's have a joint venture”
0:21:19 · David Weiss · Joint venture
David Weiss says IBM is making Promontory the centerpiece of its financial services business.
“And now it's IBM Watson and Promontory, and really making Promontory the centerpiece of IBM financial services.”
0:21:57 · David Weiss · IBM strategy · 2016