Wall Street believes artificial intelligence can replace thousands of analysts

But can investment banks, without training analysts, grow the next generation of leaders?”

But can investment banks, without training analysts, grow the next generation of leaders?

Wall Street job opportunities for college graduates are on the verge of disappearing. According to the New York Times, that’s because Wall Street executives claim that their generative AI software can do the work of analysts much faster and cheaper,
If Wall Street uses generative AI to replace analysts, it will make the same mistake it made a decade ago when they predicted that online courses would replace colleges and universities.

And if Wall Street uses generative AI to replace analysts, it will make the same mistake they made a decade ago when they predicted that online courses would replace colleges and universities.

How so? While attendance is important, students benefit immensely from the branding, networking opportunities and teamwork skills that top schools provide, — something online learning can’t.

And that’s not the case.

Starting at a Wall Street bank, such as Goldman Sachs or Morgan Stanley, — is an even more valuable branding and networking opportunity. Unlike colleges, investment banks need talented analysts who will develop their skills so they can run the bank when current executives burn out or retire.

An investment bank needs talented analysts who will develop their skills so they can run the bank when current executives burn out or retire.

Bearing this need in mind, why would banks want to replace analysts with generative AI? The job is undoubtedly a hazing ritual that causes misery. Analysts «spend nights piecing together PowerPoint presentations», the Times reports. «Punching numbers into Excel spreadsheets. Honing their language in esoteric financial documents that may never be read by another soul», —adds the Times.

Analytics are the only ones who have ever written a financial document.

Bank executives, however, tend to be more concerned about profits than the well-being of their employees. Banks are testing artificial intelligence tools that can do «in seconds the kind of work that now takes hours or entire weekends» the Times notes. Although I haven’t seen such software in action, Goldman Sachs says it is developing a tool that converts a «lengthy PowerPoint document» into a «legally savvy» initial public offering prospectus in a second, the Times reports. Executives could use such software to cut by two-thirds the number of analysts they hire and pay them far less than the $100,000 plus bonuses they get now, the Times notes.

The Times notes.

An expert thinks Wall Street is overestimating AI’s ability to cut costs

Banks may not realize those savings. «I personally think AI will actually create more work, not less, for junior analysts» — Alex Salkever, an AI expert and technology executive, told me in an April 11 email. While AI can simplify the job of quickly summarizing documents, much of the analytical work requires human judgment to avoid the reputational risk of having generative AI hallucinations appearing in front of the client.

At the same time, AI can make it easier for a client to quickly summarize documents.

Banks will need analytics to protect their brands. People aren’t just reformatting information,— noted Salkever, author of «Driverless Car Driver: How Our Technology Solutions Will Create the Future». They make judgments, check facts and make sure AI doesn’t hallucinate or confuse». No major bank wants to lose a customer because of critical errors in a pitch-based document» — Salkever added.

Salkever.

Why artificial intelligence won’t replace Wall Street analysts

Business leaders are looking for a return on their investment in generative AI. My forthcoming book «Brain Rush: How to Invest and Compete in the Real World of Generative AI» provides detailed examples of such applications. The highest-yielding applications of generative AI do the following:

• They provide a quantum leap in value, enabling significant increases in revenue and productivity, in a company’s critical business processes;
• They attract new customers and retain current customers; and
• They are difficult for competitors to replicate

.

In my opinion, the artificial intelligence software that these banks are developing may not pass these tests. While this technology may improve analyst productivity, it appears to be under broad industry development, so it will not create a competitive advantage for an individual company. Moreover, there is no reason to believe that this software will help banks attract new customers and retain them.

At the same time, there is no reason to believe that this software will help banks attract new customers and retain them.

I tried to use ChatGPT to create a Goldman Sachs PowerPoint presentation to convince clients to buy IBM stock. I give ChatGPT a D-minus for its efforts — it created virtually nothing of value. How so? On April 11, I uploaded 4 IBM annual reports to ChatGPT and asked the chatbot to prepare a PowerPoint presentation beginning with a summary that explained the reasons why IBM’s stock price would rise. I also asked the chatbot to graph IBM’s revenue by product and geography, gross profit, net income, and free cash flow with headlines explaining the reasons for these trends. Finally, I asked ChatGPT to predict these variables and explain the assumptions behind them. After several attempts, ChatGPT returned a PowerPoint presentation with two slides that badly reformatted the first few sentences from IBM’s letter to shareholders for 2023.

After several attempts, ChatGPT returned a PowerPoint presentation with two slides that badly reformatted the first few sentences from IBM’s letter to shareholders for 2023.

If Wall Street software can do in one second what I asked ChatGPT to do, it might improve analyst productivity. However, I doubt that chatbots will ever do the job of banking partners.

I doubt that chatbots will ever do the job of banking partners.