By mid-year, all of Morgan Stanley’s thousands of advisors are expected to have access to a new AI-powered chat tool.
The tool, which is already used by about 600 employees, gives advisors answers to questions like “Can you compare the investment cases for Apple, IBM and Microsoft?” and follow-ups like “What are the risks of each?” An advisor can asked what to do if a customer has a potentially valuable picture – and the knowledge tool can provide a list of steps to follow, along with the name of an internal expert who can help.
“What we’re trying to do is make every client or every financial advisor as intelligent as the most knowledgeable subject matter expert in real time,” said Jeff McMillan, head of analytics, data and innovation for Morgan Stanley Wealth Management.
Experts disagree on whether AI will destroy more jobs than it creates over time. But it is clear that AI will change the jobs of most knowledge workers, changing the skills they need and changing the staffing needs of most companies. It’s now up to business leaders to figure out how to take advantage of technology today while preparing workers for the disruption the tools represent in the medium term.
Moving too slowly can mean lost gains in productivity, customer service and—ultimately—competitiveness, similar to what happened to businesses that didn’t embrace the Internet fully or quickly enough. But at the same time, leaders must guard against the errors and biases that AI often supports and be mindful of what this means for employees.
“Almost no matter what sector you’re in, you should think of your company as an AI-first company,” said Alexandra Mousavizadeh, CEO of Evident, a startup that analyzes financial companies’ AI capabilities.
The type of AI that underpins Morgan Stanley’s advisor tool is called generative AI. It can create content — including text, images, audio and video — from information it has analyzed. In addition to answering questions, it can be used in countless other ways, such as drafting notes and emails, creating presentation slides, and summarizing long documents. Early research shows that tools built using generative AI can speed up many tasks and increase employee productivity.
Researchers at MIT and Stanford, for example, found that customer support agents equipped with an AI tool that suggested answers resolved an average of 14 percent more customer issues each hour.
But the gains were not evenly distributed. Less experienced workers made bigger leaps in productivity because the tools effectively “captured and propagated” the practices of their more skilled colleagues. Other recent MIT research similarly noted that workers who were initially not as good at tasks were able to close the gap with those who were more skilled by performing better and spending less time when they were assisted by AI
One possible conclusion from these findings is “that the advantage someone had from tenure in terms of their performance has now diminished because a young person with ChatGPT can perform as well as someone who has several years of experience,” Azim said. Azhar, chair of the Exponential View Research Group. If the research plays out in wider practice, it could potentially lead some companies to invest more in younger employees while shifting less to more expensive workers who work longer.
Some companies are already starting to make staffing decisions based on the expected impact of AI tools. IBM recently said it is slowing or stopping hiring for some back-office roles, such as human resources functions, that could be replaced by AI in the next few years.
Increasing speed and productivity from AI will raise customer expectations, said Bivek Sharma, chief technology officer for PwC’s global tax and legal services. “Then it’s about making sure we can reskill the workforce quickly enough and enable them with AI quickly enough to meet the obvious demand that’s going to come on the back of that,” he said.
PwC is working with Harvey, an AI startup that builds tools for lawyers, to roll out an AI chat tool across its legal advice practice over the next few months. It plans to extend this technology to its tax and human resources experts as well.
Besides quickly providing answers to staff members that draw on the firm’s expertise, PwC’s goal is to generate new insights, including possibly by analyzing its clients’ data, Mr. Sharma said. AI could potentially feed into all the contracts of two companies considering a merger, for example, and allow PwC experts to query specific types of provisions and risks.
“Think of it as a scale-up game, not a time-saving game for us,” Mr. Sharma said. “It’s almost like a senior associate attached to each of our legal and tax advisors, increasing what they can do every day for their clients.”
Larger companies typically need to invest in AI-savvy technical staff who can adapt the technology for their business. Already, “there are companies that can’t adopt ChatGPT because they just don’t have the main rails to run it on, which is content and data management in a row,” Ms. Mousavizadeh said.
They also need to hire or train new professionals for roles that don’t necessarily require technical knowledge. Morgan Stanley’s Mr. McMillan and other corporate executives say AI platforms require constant “tuning,” with people adjusting parameters and sources of information to get the best results for users. This setup created a need for a new group of workers known as “fast engineers” or “knowledge engineers.”
Morgan Stanley and PwC are among those building their own versions of AI chat tools that draw on internal material.
Concerns about security, privacy, accuracy and intellectual property rights have led many companies to limit their employees’ access to public ChatGPT and other generative AI tools. They want to avoid what reportedly happened at Samsung, where employees working in the semiconductor division are said to have shared confidential computer code and meeting notes while using ChatGPT. Executives are also concerned about frequent errors and built-in biases with some AI tools.
But part of the opportunity with tools that use generative AI that allows users to enter questions or commands in normal language is to involve a wider group of non-technical staff members in figuring out how it can change a company’s business. “Your people need to be using these tools really, really regularly so they can start building their competencies and your own internal company competencies,” Mr. Azhar said.
It suggests that public AI tools can be used in ways that do not compromise privacy or security. For example, an employee might ask ChatGPT about the best ways to combine types of sales data to tell a compelling story without actually entering the data itself. The opportunity, he says, comes from “frontline employees, regardless of seniority, who decide to improve their work through generative tools.”
Kevin J. Delaney is the co-founder and editor-in-chief of Charter, a media and research company focused on the future of work.