How is AI changing careers in financial research?
Equity research has changed a lot over the years. Once, it was an area obsessed with unbundling and juniorization. But, as with any other industry, there is a five-ton elephant in the room: the rise of artificial intelligence.
Click here to join the bubble by eFinancialCareers, our new anonymous community. ✍️
AI can churn through reams of text and data in a matter of minutes, if not seconds. The role of an analyst has changed; maybe for the better.
“No analyst wants to spend an afternoon digging through old notes,” said a Bloomberg research piece earlier this month, “but speed is not the main prize.”
If speed is not the main prize in research, what is? Capacity, apparently. AI can give investment professionals “room to do the work that cannot be reduced to retrieval or summarization.” For Bloomberg, this includes “weighing evidence, testing assumptions, disagreeing with each other and deciding whether the market is missing something.”
AI is very much in research already. Ratings agency S&P Global, for example, built an agentic AI-powered tool called the “Credit Memo Builder” to assist work on “manual data entry” to more fruitful pursuits, such as “analytical evaluation of borrower risk” and “strategic credit assessment.”
S&P itself says that the goal of the tool is primarily “cutting through fragmented data”. But that fragmented data might have value in and of itself. Bloomberg’s report pointed out that “the most valuable information” for an investment team was a firm’s own memory: “the analyst note that challenged consensus two years ago, the investment committee debate that killed a trade, the portfolio manager’s comments after a position went wrong, the assumptions behind a thesis that later proved right.”
AI's key research talent is its ability to take all the data that might be valuable to forming an opinion and to standardize it, essentially. In many cases, “research is scattered across emails, shared drives, chat messages, spreadsheets, PDFs and personal filing habits,” Bloomberg’s report said. AI has a bird’s eye view of absolutely everything.
In the circumstances, it's hardly surprising that low-end research jobs are disappearing. Data from market intelligence provider Substantive Research, published in May this year, showed that some 740 “mid-tier” analysts had left the research profession, which was increasing skewed towards senior people. 80% of large asset management firms polled by Substantive Research intended to “slightly” increase their budgets to better pay senior, well-regarded analysts.
There are already companies mass-producing research, such as AIBNKO, Hudson Labs and Samaya AI. AIBNKO says it creates and distributes institutional-grade equity research,” entirely without human intervention. “The platform can track and cover an unlimited number of global equities in real time,” its announcement press release said proudly, “a scale no traditional sell-side desk, constrained by analyst headcount, can match.”
Fortunately, there’s more to equity research than just feeding data to a model. When we spoke to Rothschild & Co Redburn analyst Fergus Neve, much of his day was filled with phone calls and meetings with the firms that he covers – AI can’t do that, yet. Boutique firms like Woozle Research, whose business involves humans conducting interviews with humans and eliciting qualitative information inaccessible elsewhere, might become more valuable in the future.
Brett Caughran, a former Citadel portfolio manager, said on X in May this year that many hedge funds weren’t interested in hiring juniors anymore as AI agents were “solving process bottlenecks that heretofore required the efforts of junior investment talent across modelling, meeting prep, and data analysis.”
One of AI’s most early stalwart bears – Citadel CEO Ken Griffin, who previously called it garbage – has come around to the technology. Citadel employs a lot of analysts that perform a lot of research. “Work that we would usually do with people with masters and PhDs in finance over the course of weeks or months being done by AI agents over the course of hours or days,” Griffin said back in May.
Of course, there is one limitation that AI will simply never be able to surmount: it cannot be accountable. Research notes are serious things that constitute investment advice, and banks are liable for their publication. A human being can be held accountable for a research note, and therefore a human being must sign off on a research note, and therefore a human being will want to verify what they are putting their neck on the line for.
Click here to return to "What are bank research jobs?".
Have a confidential story, tip, or comment you’d like to share? Contact: +44 7537 182250 (SMS, WhatsApp or voicemail). Telegram: @SarahButcher. Signal: sarahbutcher.22 Click here to fill in our anonymous form, or email editortips@efinancialcareers.com.
Bear with us if you leave a comment at the bottom of this article: comments are moderated intermittently by human beings. Sometimes these humans might be asleep, or away from their desks, so it may take a while for your comment to appear. You must take sole responsibility for comments you post on this site. We will take reasonable steps to weed out anything that we consider to be offensive or inappropriate.
