What if you heard about an investment adviser who never grows tired, never needs sleep, who spends every moment evaluating endless amounts of new information, discovering, learning, always learning?
You’d hire him, right?
If you ran an investment management company and a computer program was a better stock picker — and didn’t expect million-dollar compensation — you’d use it, right?
That’s the scenario facing the financial services industry, including investment or wealth managers.
Artificial intelligence — AI — software that employs “machine learning” has made huge strides in the past few years, primarily because of improvements in computer processing power and data storage. And now AI is posing a threat — or opportunity, depending on your viewpoint — to the wealth management sector.
From American Banker (“Beyond Robo-Advisers: How AI Could Rewire Wealth Management”) to Newsweek (“Goldman Sacked: How Artificial Intelligence Will Transform Wall Street”), news stories abound describing this coming transformation.
In Arkansas, however, wealth management firms told Arkansas Business that while they were aware of AI, the topic wasn’t something they wanted to discuss. Stephens Inc., Foundation Resource Management, Arvest Wealth Management, Simmons First Investment Group, Westrock Asset Management — none of them had any opinions to share.
“As you know, Arvest strives to be a technology-forward bank when it makes sense for us and our customers, and it is fair to say we’re staying abreast of the role artificial intelligence may play not just in investments, but banking as a whole, in the future,” said a spokesman for Arvest, which has $1.5 billion in assets under management. “Because it is such a broad topic and the landscape is rapidly changing, however, we are not prepared to discuss specifics at this time.”
But there will be no escaping AI. That’s because AI is what’s called a “general-purpose technology,” a technological innovation so profound that it transforms all of society. Some earlier examples: the domestication of plants and animals (agriculture), the wheel, writing, the internal combustion engine, electricity, mass production, the computer, the internet.
“The most important general-purpose technology of our era is artificial intelligence, particularly machine learning,” said the Harvard Business Review late last month, in a “Big Idea” cover story headlined “The Business of Artificial Intelligence.”
“The fact of technology advancing is inevitable and is likely to change wealth management as we know it today — I think for the better,” Bob Dannhauser, head of the Global Private Wealth Management practice at the CFA Institute in New York, told Arkansas Business last week.
But, he said, the changes may not be what people expect.
“The common conception of it today is that computers will essentially replace humans and take over a lot of the investment decision-making and investment counseling that goes into helping people manage their wealth,” Dannhauser said. “And I think that’s unlikely. At least, it’s unlikely in the way that computers would have complete autonomy over decision-making.
“I think there’s always going to be a place for human beings to offer their judgment. Indeed, I think that’s what will help distinguish effective wealth managers in the future from less effective ones.”
Already Here
AI already is all around you. Facebook is using it to improve your news feed, to show you content — news stories, ads — you find interesting or useful. Siri, the personal assistant found on Apple products, is an example of AI; so is Netflix, when it recommends movies to you based on your previous choices.
What is unique and wonderful and startling about artificial intelligence — again, depending on your viewpoint — is that it can learn; it can get smarter and smarter.
Machine learning is the process that allows this constant evolution. Instead of a computer being programmed to solve a particular problem, it is programmed to study huge amounts of data and then make predictions based on the examples provided by that data. AI, using machine learning, actually learns from history, unlike many humans.
In addition, artificial intelligence, programmed correctly, lacks the biases and assumptions that can cloud human decision-making.
“From the point of view of the investment companies, the first advantage of AI would be to create a more consistent approach due to the detachment of emotion from the investment decision and the minimization of human errors in the execution,” said Sergio Santamaria, a senior research associate at the Garrison Financial Institute at the University of Arkansas at Fayetteville. Arkansas Business interviewed Santamaria, a chartered financial analyst who also is the president of the CFA Society Arkansas, both by email and by phone.
AI differs from quantitative analysis, which uses statistical models to devise investment strategies and which has been used for years in hedge fund management.
“Artificial intelligence goes beyond manipulating data to actually making decisions about the data,” Dannhauser said. “Quantitative management has largely rested on models that ultimately help portfolio managers make decisions about which securities to buy and sell or how to structure portfolios. Artificial intelligence might be components of that decision-making taken over by computers.”
Santamaria said that AI, unlike quantitative analysis, “can develop a continuously [self-improving] system that can apply the selected strategies to more markets and instruments faster than any previous technology.”
AI also differs from robo-advisers, which collect information from clients about their finances and goals and use computer algorithms to provide investment advice.
One of the aspects of artificial intelligence that makes it so powerful is its ability to gather and evaluate information from countless sources — not just from company earnings reports, regulatory filings and historical performance records, but from news stories and social media. Anything that can be digitized can be analyzed. And AI can perform this analysis in real time.
Mark Minevich, principal founder of Going Global Ventures of New York, a global investment, technology and strategic advisory firm, used EmmaAI as an example of the power of artificial intelligence.
“EmmaAI looks to outpace day traders by analyzing a complex set of factors using a system of neural nets, which emulate elements of the human cognitive process by ‘learning’ patterns in more data than could ever possibly be handled by humans,” said Minevich, who is also a senior adviser to the Council on Competitiveness in Washington, D.C.
“Prominent companies are already using AI to aid their forecasting efforts,” he noted, citing investment management firm BlackRock of New York, which “built Aladdin, an AI platform that uses natural language processing to analyze thousands of documents and help financial managers make rapid, informed investment decisions. The platform … can do the work of thousands of analysts in a fraction of the time.”
Santamaria, at UA, said that though the adoption of artificial intelligence has so far been limited in the asset management industry, it’s poised for explosive growth.
And artificial intelligence will affect the entire financial services sector, not just wealth management. JPMorgan Chase & Co. uses a program it calls COIN — for Contract Intelligence — to review commercial loan agreements. Work that took lawyers and loan officers 360,000 hours can be done by COIN in seconds.
That leads to another consideration: labor costs. “Given that investment professionals are generally well compensated, the trend will be to replace financial workers with algorithms in order to save money for the investment firms and consequently for the clients,” Santamaria said.
Less Tangible Benefits
The CFA Institute’s Dannhauser said AI could provide other, less tangible benefits to investment management — for example, reserving for humans the tasks that they do best rather than less creative, rote tasks.
“If you think back 20 years ago, a lot of people in our industry were still making decisions about investments on the basis of manual models,” he said. “They’d take their green ledger sheets and they’d track revenues and expenses for companies and make some conclusions about trends. And then spreadsheets were introduced and that automated and made the power of analysis much more available to many more people and transformed the way security analysts and portfolio managers did their jobs.”
Despite AI’s inevitable inroads, it’s unlikely to replace human investment advisers, Santamaria and Dannhauser said. Santamaria compared the situation to the airline industry, “which relies heavily on autopilots and also uses human pilots during riskier maneuvers such as takeoff and landing.”
Asked if artificial intelligence posed an existential threat, Dannhauser was reassuring. “I don’t think investment advisers should feel particularly threatened in that regard, to the extent that they’re able to adapt and make the best use of technology as it comes on the scene.
“For the ones who are sort of hard-nosed and say, ‘The old way is the best way. I’m never, ever going to change my approach,’ they may actually be OK too, because I think there’s probably a segment of clients out there that will also appreciate that approach.
“But there’s a little bit of a danger, I think, to being left behind or not taking best advantage of the resources that you have available to make smart decisions for your clients.”
Sign up for Daily E-news
