How Morgan Stanley Is Optimizing AI for Client Growth

Due this spring is the same capability for mobile devices and designed for in-person meetings. Morgan Stanley has another 75 or so AI ideas in the hopper, Cucchiara notes.

A major step in the firm’s journey to digitize its process, he explains, was the September 2023 launch of “The Assistant,” an AI chatbot giving quick access to 100,000 of Morgan Stanley’s documents, or “intellectual capital.”

In the interview with Cucchiara, who joined Morgan Stanley in 2016 after 20 years with Bank of America Merrill Lynch, he discusses the wirehouse’s experimentation with AI to allow advisors to ask questions and receive on-screen answers. Then the technology can provide names of clients who may benefit from the specific subject matter.

Here are excerpts from our conversation:

THINKADVISOR: What’s your goal in providing artificial intelligence to your advisors?

SAL CUCCHIARA: We’re focused mostly on raising productivity: How we can take the administrative nature out of the job to allow advisors and service teams to service clients better.

But is the use of AI improving advisors’ bottom line?

We’re freeing up their time. When you have more time, the goal is, hopefully, to serve more clients. That’s why we’re not looking at this as a cost-reduction exercise. It’s more of a growth exercise.

That’s how we’ve positioned it: Here to help.

But by using AI, isn’t there less need for sales assistants and administrative assistants?

This is not an opportunity to reduce the size of our team. It’s a productivity enhancer to service clients with more value-added activities.

It will help the service assistants focus on that.

When do you think AI will be able to strategize like humans?

Everybody has thought that advisors can be displaced, originally, by online brokerage, then by robos [digital advisors].

But I don’t see AI disintermediating the client and their advisor. AI will enable advisors to focus on serving clients. If that means serving more clients, great. But it’s really focused on servicing the clients they have.

How will AI reshape wealth management?

Within the framework we have, AI is the next generation of digitizing our process.

We’ve digitized all the information that’s structured data. Now we can digitize all the information that’s in an unstructured format.

Bringing those two worlds together is the next leap in making things easier to do, how we take more administrative things out of the hands of our workforce. That’s really the key.

What AI system do you use?

OpenAI and GPT-4 right now. We continue to look at where the AI industry is going. We have a lot of relationships across other large language model vendors.

For the things we’re trying to solve in wealth, OpenAI works very, very well.

Are you still using source material that’s exclusively from Morgan Stanley?

That’s the beauty of our AI development. We have a secure environment: our own infrastructure.

We didn’t want to confuse anything about our procedures or policies with anything the [large language models] may have learned scanning the Internet.

Open AI doesn’t learn anything about our private information or our client information. They’re not training their large language models on our information. They don’t retain any of it.

What’s your newest AI rollout?

It’s called Debrief. With the permission of the client, we can record an advisor’s [video conference] meeting with them and then summarize it.

We give the advisor a sample email that they can edit, save and send to the client.

So the advisor is still in the middle and taking the next steps. It ensures that the advisor is the [center] of anything we do with AI.

In the next quarter, we’ll be rolling out [that capability] so that it’s on the advisor’s mobile device, and they can bring it to in-person client meetings.

What else do you have under development using AI?

To be rolled out at the end of this year is creating agents [customized AI “co-pilots”] for our client service teams, the folks who take calls from clients.

With AI, we can give them very quick access to information that’s at their fingertips. They can answer questions on the spot to help the client in whatever they’re trying to do.

Any additional AI applications you’d like to discuss?

For the hundreds of thousands of documents we receive in foreign languages — because we have a global client base — we can translate and summarize them for putting plans together for prospective clients. Our experts create the actual financial plans.

We’re using AI for translating statements, say. This is a way to [expedite] extracting information from a statement and allowing our financial planners to use it.

How much training is involved for the advisors to learn AI-enabled processes?

A little bit of light training is required.

What are some of the most complex AI uses you’re developing now?

We’re working on how to use AI to make our platform more conversational. That strategy is in the works, and we’ll get there by the end of the year.

[For example], we’re experimenting with the ability for the advisors to ask questions like, “How did Microsoft do in earnings?” or “What’s our latest research report?” And AI will pull that up on their screen.

The advisor can then ask, “Which of my clients holds Microsoft?”

Your AI rollout in September of 2023 was a chatbot dubbed “The Assistant,” which provides 100,000 Morgan Stanley documents on policy, procedures and other how-to applications. How was advisor acceptance?

Highly successful. We have an over 98% adoption of that tool. You can search, and it can summarize things, not just find them. Advisors can interact by asking questions.

It’s raising the overall productivity of advisors when they have to search for a policy, a procedure, things around investments. That was a good building block for us.

Do you talk much about using AI to advisors who are new to the firm?

Yes, when we bring them in. AI helps them because they can ask The Assistant how to do something and get the answer right away.

AI-enabled information makes it very easy for an advisor, especially a new one. We see that as a big win.