With some fearing that AI will make many white-collar workers' jobs redundant, at least one new job title exists that didn't before: chief artificial intelligence officer.
Feld — who will continue to serve as the firm's head of front-office technology — was previously a managing director at JPMorgan, where he led global teams supporting wealth management, audit and corporate technology. Before that, he was a managing director at UBS, leading wealth management investment solutions and banking technology teams, and a software engineer at Prudential Securities, where he held multiple leadership positions in the technology organization, including divisional chief investment officer at Prudential Financial.
READ MORE:
In explaining his newly created role in an interview with Financial Planning, Feld referenced the
"Clearly, this is not a fad, but too often people get very excited for five minutes like with blockchain," he told Financial Planning. "It's still around. It's still an important technology, but it didn't change the world for us. There have been a lot of these that have come and gone over the years."
Feld said he is "100% sure AI has legs."
"But you have to be careful not to dive into any technology full speed ahead," he said. "Because, one, it could turn out to be hype and you may not realize the gains you think. And, two, something that's moving as fast as this, whatever we do today, is almost guaranteed to be obsolete 12 months from now."
READ MORE:
Feld also took the time to speak with Financial Planning about his vision for his new role, how
This interview has been lightly edited for clarity and length.
Financial Planning: At Raymond James
Stuart M. Feld: My understanding is clear. It's, 'how do we leverage AI in a responsible and productive way?' We make sure we understand what those models do — what they're supposed to be — by continuously monitoring them. We make sure people understand how to use this technology. It doesn't help us to have this technology if business partners don't know how to leverage it. The vision is to make sure that we can realize this technology and then have the tech to build it.
FP: What is an example of how you are already utilizing AI in your tech stack?
SF: What we're doing with Zoom AI. We want to be able to summarize conversations that advisors are having with clients, so that saves them a ton of time, and we are able to directly input that into our CRM system. Then any tasks or appointments that come up in that meeting can automatically be applied to our homegrown CRM system. … The biggest thing we hear from advisors is, 'Save me time so I can service my clients.'
Also, like every big firm,
We've begun a pilot to AI-power our intranet. Like all firms, there's a lot of data there, and it was a keyword search. Now you can ask questions and get natural language responses. Before you had to pick up the phone and call a help desk or ask somebody else. We got a great email from an advisor saying, 'I just tried this new pilot search tool and something that would be very hard to get, usually would be very hard to find out the information, I got in two seconds just by asking it a question.' That was great to see. That's starting to pay dividends rolling out now. Everyone in the firm will have it by the end of the second quarter.
FP: What about using AI in something like financial planning? Have you explored that?
SF: We have good financial planning tools. For now, that's not a space where we want to go. Certainly, we could use it, but we need to make sure we know exactly what the black box is doing. We definitely need more control. We don't want to just send that into the gen AI space and say, 'Come out with a plan for this person,' and hope that it's given us the perfect plan. I don't think it's there yet. I think it will get there, but we have to prove that the model is behaving the way we want it to behave.
FP: How can people feel safe with AI having access to such personally identifiable information (PII)?
SF: I'll stop you there. Security is first and foremost for us. We don't have anything that's open, that can be trained on, none of our information is. All of our information is encrypted. We have contracts with the vendors that we use that they cannot train and even retain information that is in their systems. That's less of a fear for us. If you go into ChatGPT, that's not going to the same place where Raymond James goes in. We have our own private areas.
FP: Do you have any fears regarding the effect AI will have on the wealth management industry?
SF: That's kind of like the robo conversation that we had. Five years ago, people were concerned. 'Well, what do we need an advisor for? We have the robo advisors.' It's the same thing here. It's just a technology. It's not making decisions. It's not replacing humans at this point. From the advisors I speak with, most are not concerned about that. We're certainly not trying to use the technology in any way, shape, fashion or form to replace an advisor. That's counterproductive for us.
FP: What about your greatest hopes for AI?
SF: The next phase will be to tackle more interesting subjects. Right now, we're automating things in the back office. But I think as it progresses, we'll start to get to help with portfolio construction. And that's not black box, but that's inputting all the data about the client, mixing financial planning data, risk tolerance data and market data, and coming up with a better way to build portfolios. That's the next big thing, and that's going to take a little more work on our part. We have plenty of work to do on the administrative side. Again, not replace —we need to be crystal clear about that. We're just doing it at lightning speed.
FP: There was a shakeup in the AI sector recently when
SF: We're not building large language models (LLMs) here. We're using other people's LLMs. That was their big accomplishment, building an LLM for what they claim is a fraction of the cost. We're not in that business, and so it wouldn't impact us. And we're not looking at them at all. We would only do business with established companies like Microsoft and Google. It wouldn't benefit us.
