Morgan Stanley is testing OpenAI’s chatbot technologies for its economic advisors. Mike Kemp—In Photos/Getty Pictures

OpenAI’s chatbot, ChatGPT, had gained millions of customers for its capacity to produce extended prose primarily based on brief prompts. 

Regardless of its fast adoption, the chatbot is not without having its challenges. The tool can occasionally make errors, or “hallucinate,” analysts at investment bank Morgan Stanley mentioned in a note final month. At the time, the analysts wrote that ChatGPT can “generate answers that are seemingly convincing, but are truly incorrect.” 

However that hasn’t stopped Morgan Stanley from testing an OpenAI-powered chatbot with its economic advisors. The objective is to assistance the advisors make the most of the bank’s big library of investigation and information sources, the firm announced Tuesday.

The experiment is aimed at “helping investment experts parse by way of thousands of pages of our in-depth intellectual capital, analyst commentary, and market place investigation in seconds – a course of action that usually could take additional than half an hour,” Morgan Stanley Wealth Management’s head of analytics and information, Jeff McMillan told Fortune. “This will assistance advisors invest additional time focusing on serving their clientele.”

The tool will use the technology’s newest version, GPT-four, which launched on Tuesday. It is presently becoming tested with 300 advisors, CNBC reported, and when released additional broadly, will help all of Morgan Stanley’s 16,000 advisors.

The move marks the very first identified use of ChatGPT in the banking sector, which has faced a tumultuous couple of days amid the failure of Silicon Valley Bank. In the economic technologies space, Stripe, an on-line payments processing firm, is also testing a version of ChatGPT to fight fraud. 

Morgan Stanley is no stranger to making use of artificial intelligence. It presently makes use of it to realize and cater to client requirements by matching them with the suitable economic advisors.

The bank’s use of GPT-four will be intended to help human advisors, a lot like a investigation assistant would. But it is not supposed to replace human advisors, who are nonetheless required to interact with clientele. 

“These issues (A.I. tools) do not have any empathy they’re just really clever math that is capable to regurgitate know-how,” McMillan told CNBC.

Even even though Morgan Stanley’s analysts acknowledged the ChatGPT’s struggle with accuracy, they didn’t create off the technologies completely. In a February note, they wrote that the “A.I. hype is worth taking into consideration seriously” and that it had “real market place prospective.”

McMillan mentioned that making use of ChatGPT in banks comes with inevitable complications. For instance, the banking organization is extremely regulated and includes the important function of handling people’s funds. 

“What tends to make the function we’re carrying out especially intriguing is that it is no little feat to integrate technologies into a extremely complicated and regulated atmosphere like ours, and to do so with the suitable controls installed,” McMillan told Fortune. 

OpenAI did not right away return Fortune’s request for comment.

Subscribe to unlock this report and get complete access to Fortune.com

Attempt for $1

Currently a member?
Login

OpenAI’s chatbot, ChatGPT, had gained millions of customers for its capacity to produce extended prose primarily based on brief prompts. 

Regardless of its fast adoption, the chatbot is not without having its challenges. The tool can occasionally make errors, or “hallucinate,” analysts at investment bank Morgan Stanley mentioned in a note final month. At the time, the analysts wrote that ChatGPT can “generate answers that are seemingly convincing, but are truly incorrect.” 

However that hasn’t stopped Morgan Stanley from testing an OpenAI-powered chatbot with its economic advisors. The objective is to assistance the advisors make the most of the bank’s big library of investigation and information sources, the firm announced Tuesday.

The experiment is aimed at “helping investment experts parse by way of thousands of pages of our in-depth intellectual capital, analyst commentary, and market place investigation in seconds – a course of action that usually could take additional than half an hour,” Morgan Stanley Wealth Management’s head of analytics and information, Jeff McMillan told Fortune. “This will assistance advisors invest additional time focusing on serving their clientele.”

The tool will use the technology’s newest version, GPT-four, which launched on Tuesday. It is presently becoming tested with 300 advisors, CNBC reported, and when released additional broadly, will help all of Morgan Stanley’s 16,000 advisors.

The move marks the very first identified use of ChatGPT in the banking sector, which has faced a tumultuous couple of days amid the failure of Silicon Valley Bank. In the economic technologies space, Stripe, an on-line payments processing firm, is also testing a version of ChatGPT to fight fraud. 

Morgan Stanley is no stranger to making use of artificial intelligence. It presently makes use of it to realize and cater to client requirements by matching them with the suitable economic advisors.

The bank’s use of GPT-four will be intended to help human advisors, a lot like a investigation assistant would. But it is not supposed to replace human advisors, who are nonetheless required to interact with clientele. 

“These issues (A.I. tools) do not have any empathy they’re just really clever math that is capable to regurgitate know-how,” McMillan told CNBC.

Even even though Morgan Stanley’s analysts acknowledged the ChatGPT’s struggle with accuracy, they didn’t create off the technologies completely. In a February note, they wrote that the “A.I. hype is worth taking into consideration seriously” and that it had “real market place prospective.”

McMillan mentioned that making use of ChatGPT in banks comes with inevitable complications. For instance, the banking organization is extremely regulated and includes the important function of handling people’s funds. 

“What tends to make the function we’re carrying out especially intriguing is that it is no little feat to integrate technologies into a extremely complicated and regulated atmosphere like ours, and to do so with the suitable controls installed,” McMillan told Fortune. 

OpenAI did not right away return Fortune’s request for comment.