“We’re probably in the second or third innings.”
It is at Andrew Lo progress report on advances in artificial intelligence (AI), big data and machine learning applications in finance.
Lo, professor of finance at MIT Sloan School of Management, and Ajay Agrawal of the Rotman School of Management at the University of Toronto shared their views at the first CFA Institute Alpha Summit in May. In a conversation moderated by Mary Childsthey focused on three main concepts that they believe will shape the future of AI and big data.
Lo said applying machine learning to areas like consumer credit risk management is definitely the first inning. But the industry is now trying to use machine learning tools to better understand human behavior.
In this process, the big question is whether machine learning will ultimately amplify all of our existing human biases. For his part, Agrawal does not think so.
“If we had had this conversation a few years ago, the issue of bias wouldn’t even have come up,” he said. “Everyone cared about training their models. Now that we’ve reached utility in a number of applications, we’ve started to worry about things like bias.
So where does the concern about bias come from?
“We train our models from various types of human data,” Agrawal explained. “So if there’s a bias in human data, the AI not only learns the bias, but it can potentially amplify the bias if it thinks that will increase its ability to optimize or make better predictions efficiently.”
But AI can also be used to minimize bias. Agrawal cited a University of Chicago study in which researchers developed AI programs that not only mimicked the bail decisions of human judges, but also predicted flight risk more accurately.
2. Economy and distribution of wealth
There is no doubt that AI increases productivity. But will AI cause a jobs crisis by making human workers obsolete? According to Agrawal, people are worried because we don’t know where the new jobs will come from or whether those who lose their jobs later in their careers will be able to retrain for these new positions.
Innovation is happening so quickly today that we don’t know if retraining programs will be as effective as they have been in the past, even for young workers who have the time and flexibility to really participate.
The other problem is the distribution of wealth. Will the adoption of AI lead to greater concentration of wealth?
“I would say that almost all economists agree with the idea that this will definitely lead to economic growth, and therefore to the overall increase in wealth for society,” Agrawal said. “But there is a division among economists as to what this means for distribution. Some of us are very worried about distribution.
There are plenty of opportunities in the financial industry for new types of data, according to Lo.
“There’s so much more we need to understand about the financial ecosystem, especially how (inputs) interact with each other over time in a stochastic environment,” he said. “Machine learning is able to use large amounts of data to identify relationships that we weren’t currently aware of, so I think you’re going to see much faster progress from all of these AI methods that have been applied to a much smaller data set so far.
Agrawal raised a related concern: “In regulated industries such as finance, healthcare and transportation, the barrier for many of them is not data. We are prevented from deploying them due to regulatory hurdles.
Lo agreed on the possibility of regulations hindering progress.
“There’s a complex set of issues that we don’t really know how to regulate right now,” he said. “A good example is autonomous vehicles. Currently, the laws are set so that if someone is involved in an accident and kills another passenger or pedestrian, they are responsible. But if an AI is responsible for a death, well, who is responsible? Until we resolve this aspect of the regulations, we will not be able to make the kind of progress that we could. »
AI and machine learning for everyone
So how can finance professionals develop skills in machine learning, big data and artificial intelligence?
“There are a lot of really, really helpful courses that you can actually take to get up to speed in those areas,” Lo said. “But it just takes some time, effort and interest to do it.”
The younger generation is best placed in this regard, according to Lo. Indeed, young people today trust human-machine relationships more, Agrawal said, because they’ve simply had more time to spend on computers, mobile devices, and more.
As Lo explained at the beginning, we are still at the beginning of applying these new technologies to finance. There are high hopes that they will boost productivity and lead to greater profits mixed with concern about the potential ramifications for the concentration of wealth and employment.
Nevertheless, concerns about the adoption of AI and big data amplifying human biases may be overstated, while the potential barriers posed by regulation may be understated.
Yet given the inevitable adoption of AI in finance and beyond, finance professionals can’t afford not to know.
If you liked this article, don’t forget to subscribe to the Enterprising investor.
All posts are the opinion of the author. As such, they should not be construed as investment advice, and the opinions expressed do not necessarily reflect the views of the CFA Institute or the author’s employer.
Professional Learning for CFA Institute Members
CFA Institute members are empowered to self-determine and report professional learning (PL) credits earned, including content on Enterprising investor. Members can easily register credits using their HGV tracker online.