What makes a good financial advisor?
Ask a simple question, but answer a complex one.
You’d imagine it would take someone with years of real life experience to give an appropriate response. Someone who’s been in the job long enough, or has worked with a consultant long enough, to understand the difference between good and bad.
But what if the answer came from an entity that was only a few weeks old, lived on the other side of the screen, and had instant access to one of the deepest knowledge bases in the industry?
Meet “Mo,” Morningstar’s generative AI chatbot, which entered beta late last week and is now available on Morningstar’s Investor, Research Portal, Direct and Advisor Workstation platforms.
Mo is powered by Morningstar Intelligence Engine, a platform that combines Morningstar’s investment research library with the same Microsoft Azure OpenAI service that powers ChatGPT.
read: ChatGPT: Miracle, Meme, or Threat?
Debuting at the Morningstar Investing Conference in Chicago last month alongside company CEO Kunal Kapoor, Mo dressed in classic smart-casual A white-collared shirt poked out from under a branded corporate sweater.
Mo is a special guest during Kapoor’s 25-minute keynoteas Kapoor posed a series of questions to the AI, two Morningstar representatives—one real and one virtual—discussed around the topic on stage, and the AI happily and confidently kept pace.
“What should financial advisors tell potential clients about finding the right strategy among hundreds of thousands of investment options?
“How does Morningstar rate the ARK Innovation ETF, ARKK?”
“Which investors are suitable for the tax benefits of direct indexation?”
But the head of the Chicago-based firm known for its fund rankings, investment data and market research asked this very important question, back and forth.
“What does it take to be a good financial advisor?” Kapoor asked Mo.
“A good financial advisor should provide comprehensive financial and portfolio planning and have a wide range of expertise, including tax experts, investors and estate planners,” Mo responded. “They should also have a certified financial planner credential and be able to add value to clients through their skills and advice. However, the level of fees and the investor’s own skill level should also be considered.
“Ultimately, a good financial advisor prioritizes understanding the client’s goals and intentions and providing personalized advice and solutions.”
The answer Mo gives is lighthearted and flattering, but some word choices suggest it’s a work in process, as far as utter naturalness goes.
The vibe is very much like Jake from State Farm, but the insight is unique to Morningstar. And for good reason.
When a user types a question into Mo, that question is routed to an engine built on the Morningstar database, which stores embeds from Morningstar content.
So when Mo receives a question from a user, the Morningstar Intelligence Engine identifies the most responsive content to feed to OpenAI’s large language model and builds the response. Responses are then tested for relevance and responsiveness before being fed back to users.
Morningstar officially chose to cooperate with Microsoft’s Azure OpenAI Service because Microsoft promised not to allow user data to train OpenAI’s large language model. In keeping with Morningstar’s commitment to privacy, users are advised not to enter personal or confidential information into Mo.
James Rhodes, Chief Technology Officer and President, Data, Research and Enterprise Solutions, Morningstar financial plan In an interview, Mo’s ability to forget and recall effortlessly deserves a mention.
But Mo’s poor memory from his vast knowledge is a deliberate character flaw.
“Every chat you have with it is new. The reason we’re doing it this way is actually for security and privacy. So we don’t tie issues back to users,” Rhodes explained. “We want people to feel comfortable asking all types of questions. We don’t want users to think we’re tracking the fact that (someone) asked X, Y, and Z. That information only comes in when you ask it, and then it disappears .”
It’s a surprising approach, considering the current software landscape is rife with apps that pride themselves on remembering every little thing as a way of delivering a more powerful experience.
Instead of tracking users and the questions they ask, Morningstar closely monitors the replies Mo gives to verify responsiveness and accuracy, Rhodes said. The company believes that Mo’s ability to provide rapid insights is needed now more than ever.
Investors are faced with a deluge of information and an “ever-growing ocean” of more than 750,000 investment options, officials said. They believe that large language models and other AI tools can make it easier to break down all these options.
Morningstar customers can ask questions directly to Mo and receive a concise answer within seconds, rather than using keyword searches and results pages that may not have been properly vetted.
However, Rhodes also made it clear that “I don’t know how to manage money.” The company made it clear that it does not provide investment advice, don’t ask.
Nor does he plan to pursue a position as a financial advisor. So feel free to invite him to your next client meeting.
“Mo is not automated investment advice. Mo is a tool that allows advisors to quickly summarize content and research ideas, making them more effective and efficient in their work,” he said. “But we have no intention of building a ‘robo-Mo’.”
read: Why the Metaverse Is Still a Big Possibility for Wealth Management
At the Morningstar conference, Mo’s digital avatar and voice recognition reps were warmly received and heating up, Rhodes said. Participants asked Mo more than 1,000 questions about investment terms, retirement savings, Morningstar’s ratings on specific securities, Morningstar’s views on emerging asset classes, and various market outlooks.
and when talking financial plan Rhodes said Friday that he believed the number of issues submitted to Mo so far may have exceeded 10,000.
“You can even see an increase in clicks per user and questions per user. I know we’ve only had a week to work on it…and it’s actually going through the investor product It’s pretty evenly split between individual investors and professional products,” Rhodes said. “The level of interest in understanding the technology and the product is actually pretty incredible and dominating the conversation around the future of space, how we see technology evolving and a general interest in how systems are set up and how they work.”
Rhodes also sees great potential for AI to support a better investor experience at scale, but we’re really just beginning to scratch the surface of the possibilities.
Mo is the first of several applications planned to be built on the Morningstar Intelligence Engine platform, and Mo himself will grow over time.
After all, he was only five weeks old. Imagine what he’ll be doing when he turns 10 weeks.
But Rhodes is not in a hurry to see Mo grow up too fast. He re-emphasized the responsibility and safety aspects of developing new technologies rooted in generative artificial intelligence, aiming to provide a near-human experience.
This is the focus of many people’s minds AI Pioneers After AI Pioneers Step into the spotlight and warn about the tools they offer if they fall within the purview of hasty innovators.
“I’ve seen a lot of things try to exploit the hype cycle. And I’m not sure everyone is doing it in a responsible way. From my point of view, it’s very important that it’s not just about rushing to get your say out ,” Rhodes said. “You need to think about your customers, you need to think about their data, and the security and privacy of that data. You need to implement all of this with a great sense of responsibility.”