Among all the perceived threats posed by AI, JPMorgan is holding out hope that the burgeoning technology will get rid of "no-joy work" for advisors and other employees.
Speaking in New York on Monday at the megabank's annual Investor Day, JPMorgan wealth and asset management head Mary Callahan Erdoes said her firm is using artificial intelligence and similar technologies in several ways. One way is to eliminate what she called "rote, repeatable" tasks.
"We are getting rid of it at every front, and it's really important," she said. "Some of the analysts have said already they get two to four hours back in their day. And we haven't even begun again to get that out across all of the forces."
JPMorgan reported at its annual investor meeting that it plans to spend roughly $17 billion on technology in 2024, up from $15.5 billion last year. The money is being used not only for AI but also to make improvements to existing systems. Of the $17 billion in spending planned for this year, $2 billion is to go to the asset and wealth management division.
READ MORE:
In AI we trust? The peril and potential of embracing AI in wealth
How wealth managers are responding to the impact of AI in the industry
FINRA, SEC send warning on deepfakes amidst their own AI plans
JPMorgan is working with U.S. regulators on generative AI pilot projects
Morgan Stanley gives AI even bigger role with new appointment
Erdoes said another way JPMorgan is putting AI to use is by eliminating the need for employees to spend time doing what she called "hunting and pecking." She gave the example of an advisor fielding a question from a client about JPMorgan's large cap growth fund.
This hypothetical client, Erdoes said, is already invested in a rival product and wants to see a comparison of the two funds' performance. Now that request can be fed directly into JPMorgan's AI system to produce almost instantaneous results.
"Just imagine the time that saves from the callbacks, phone tag and trying to get the client going and finding all the information," Erdoes said. "So, it's endlessly helpful."
JPMorgan is just one of many large wealth management firms that are turning to AI for help with controlling costs and serving clients more efficiently. Citi, Goldman Sachs and Morgan Stanley have all announced similar plans for the technology.
Morgan Stanley, for instance, formed a partnership with OpenAI — the developer of the ChatGPT large language model — in early 2023 and announced the appointment of Jeff McMillan as firmwide head of artificial intelligence about a year later. Morgan Stanley is already using an internal system called AI@MS for tasks like drawing up summaries of advisors' meetings with clients and providing faster access to the firm's voluminous market research.
JPMorgan is thinking along much the same lines. It appointed its own AI chief in June last year, naming Teresa Heitsenrether the head of its new data and analytics unit.
Erdoes said AI has allowed the firm to essentially digitally map the expertise of Michael Cembalest, the chairman of market and investment strategy for its asset management division.
Now you can get at least some of the benefit of his "20 years of eyes on the market" through a simple computer query, she said.
"You can say, 'What does he think about energy or, wait, what about last week's agricultural piece?'" Erdoes said. "And then you can tie it with our stock-picking list — which stocks would that lead you to want to buy? — all while you're on the phone with the client."
Joel Bruckenstein, the president of Technology Tools for Today, or T3, said it's one thing for wealth managers to use AI to help with customer service or investment research. He has yet to see many firms turn to the technology for actual portfolio construction.
Regulators, Bruckenstein said, generally want advisors to be able to explain why they made certain recommendations. The trouble with so many AI systems, he said, is that no one really understands why they provide the responses that they do.
That's why, Bruckenstein said, many wealth managers will confine themselves to using things like large language models to extract data from trusted sources.
"It's one thing if we're using ChatGPT and we don't know exactly where they are getting the answers from," he said. "But if you ask a question to a Morningstar or Bloomberg database, at least we know where that data is from and that it's been vetted. So that's a different story."
Erdoes said she expects AI to not only eliminate tedious tasks but also boost income. She pointed to a slide showing that sales revenue the technology helped bring in — by doing things such as providing market and performance data or assisting with advisors' recommendations — increased by 188% from 2022 to 2023. JPMorgan has projected at a previous Investor Day $1.5 billion in savings and additional revenue from its use of AI.
JPMorgan CEO Jamie Dimon acknowledged during an Investor Day question-and-answer session that the increased adoption of AI is unlikely to come without at least a little bit of pain. If some employees find their work becomes less tedious, others might see their jobs disappear altogether.
"I think it's going to change every job, like every job that we know," Dimon said. "Sometimes it's just going to be a copilot helping you do your job. Some of it may eliminate some jobs. Some, it may create additional jobs. But you can't envision one app, one database, one job where it's not going to help aid or abet."
