University professors discuss artificial intelligence’s influence on the financial world

Sunday, January 26, 2020 - 7:23pm

Professors Michael Wellman and Rada Mihalcea discuss the role of artificial intelligence in finance as part of the Friday Night AI series at the Ann Arbor Downtown Library Friday evening.

Professors Michael Wellman and Rada Mihalcea discuss the role of artificial intelligence in finance as part of the Friday Night AI series at the Ann Arbor Downtown Library Friday evening. Buy this photo
Danyel Tharakan/Daily

About 50 Ann Arbor residents and University of Michigan students gathered in the Ann Arbor District Library to learn about the future of artificial intelligence on Friday night. 

The talk, titled “Artificial Intelligence and Finance,” featured Engineering professors Rada Mihalcea and Michael Wellman, who discussed how AI will affect trading in the financial world. 

Mihalcea started the conversation by asking Wellman about his interest in AI and what keeps him in the field. 

“Even in the 1980s, it was clear to me that the future of the world was going to be the future of intelligent machines,” Wellman said. “It was pretty inevitable that computers would eventually be able to do some of the things that people do, and maybe even do it better.” 

According to Wellman, the 2008 financial crisis demonstrates why AI is important when it comes to finance. He said AI would not repeat the same problems experienced a little more than a decade ago because algorithms can forego human error. 

“We know what happens when a financial system does not work,” Wellman said. “In 2008, we experienced that when things don’t quite match up in the financial system, the whole economy goes for a tailspin. The financial crisis of 2008 — we lost trillions of dollars of productivity, even though there was no natural disaster, no resources were destroyed … it was just a miscoordination of decisions that people were making.” 

Wellman said many companies may have already started algorithmic trading, which is when an automated trading system uses computer programs hardwired with a specific set of instructions to place a trade. During this process, decisions are made very quickly. 

“We talk about decisions being made in the blink of an eye,” Wellman said. “Well, the blink of an eye is 300 milliseconds, so computers can make dozens or hundreds of decisions back and forth in that time per trade and that is a key reason why it’s been inevitable that computers would take over a lot of activity because they can react to information so much faster than people can.” 

Mihalcea asked Wellman how the trading systems will be regulated and if they have any human involvement to make sure they adhere to the law. Wellman said though there are laws in place, there are no audit systems to make sure these systems are behaving in a legal manner. Therefore, he said algorithms may still be able to find loopholes.

“Certainly in regulated areas, including markets or credit and lending, decisions made by algorithms are still subject under the law,” Wellman said. “The laws are written under the presumption that people are making decisions. We lack, I’m afraid, thoughtfully designed audit systems and other regulatory ways ensuring that we really do understand how decisions are made.” 

For the remaining time, the audience asked Wellman questions. One individual asked what the future of AI will look like when it comes to people searching for jobs. Wellman said he believes people will still be able to find jobs in the market, though AI will play a larger part in trading in the future. 

“The substrate of trading will be algorithmic and people will be involved in tweaking the strategies and coming up with new ideas and putting them in there,” Wellman said. “But it’s not going to ever revert back to mainly manual trading.”

LSA sophomores Madison Caldwell and Reagan Miller are computer science majors with an interest in AI. Miller said he thought this event would help further his knowledge as he progresses in his academic career. 

“I haven’t been to one of these before, it’s definitely a first time,” he said. “We’re both sophomores so we’re kind of getting out there right about now, especially since we are getting more into the upper-level classes. It’s cool to see some of the things that are going to be happening and impacting us in general.”

Caldwell said she thought the event was especially interesting because it offered a real-world perspective on AI.

“It was interesting to see it from a community perspective instead of just a classroom setting,” she said. “I feel like people bring different perspectives and some of the questions that people asked I would never have thought of.”