(Interesting Engineering) - Multinational investment management and financial services company Morgan Stanley is deploying a sophisticated chatbot to support the bank's army of financial advisors powered by the most recent OpenAI technology, according to CNBC.
According to Jeff McMillan, head of analytics, data, and innovation at the company's wealth management division, the bank has tested the artificial intelligence tool with 300 advisers. It intends to make it broadly available in the upcoming months.
The tool's goal is to help the bank's advisors access its data
Morgan Stanley is one of the first financial incumbents to declare a move following the success of OpenAI's ChatGPT, which went viral late last year by producing human-sounding responses to inquiries. With more than $4.2 trillion in customer assets, the bank is a titan in the wealth management industry. The potential and dangers of artificial intelligence have been discussed for years, but general users didn't grasp the implications of the technology until after ChatGPT.
According to McMillan, the tool's goal, which has been under development for the past year, is to make it easier for the bank's 16,000 or so advisors to access its vast collection of research and data.
According to McMillan, individuals aim to be as knowledgeable as the brightest employee at Morgan Stanley. “This is like having our chief strategy officer sitting next to you when you’re on the phone with a client,” he added.
Users have been astounded by generative AI, which has spawned a race among tech titans to create new products and dazzle users, but some users have been led down unexpected pathways. ChatGPT occasionally "hallucinates and can create responses that are superficially compelling, but are actually wrong," Morgan Stanley analysts noted in a report last month.
The program will provide advisors with immediate answers, just like ChatGPT does. Nevertheless, it is built on GPT 4, a more advanced version of the technology that powers ChatGPT.
Additionally, this program exclusively creates responses based on the roughly 100,000 pieces of research that Morgan Stanley has approved for this usage, which should reduce inaccuracies. According to him, the bank employs people to verify the accuracy of responses to further eliminate errors.
“We’re trying to actually break the platform” through human testing, he said. “With high-quality information, the better models, and an ongoing monitoring process,” the bank is confident in its new tool, he added.
The action builds on past McMillan initiatives, such as deploying machine-learning algorithms in 2018 that encourage advisers to contact clients or take other steps. Knowledge workers worry that as technology advances, it will eventually be able to replace people completely.
“I think every industry is going to be in some way disrupted for what I’ll describe as routine, basic tasks,” McMillan said.
Yet, he asserted that when it comes to serving intelligent consumers, machines cannot take the position of people because of the lack of empathy.
By Can Emir
March 15, 2023
