In May of 2024, Bank of America made headlines when its junior investment banker and former Green Beret Leo Lukenas suddenly died of a blood clot. The death of Lukenas was suspected by many to have been influenced by immense stress of the job’s demands and working multiple 120+ hour weeks for Bank of America in New York City. The death allegedly sparked an attempt by some junior bankers to strike (which never materialized), with their demands generating forum discussions across Twitter and many work-related subreddits. Among the demands was a maximum hour cap of 100 hours per week, which had many people in these forums asking “What do investment bankers actually do?”
This question parallels the perhaps more notorious question of, “What does a consultant even do?” which is answered both realistically and satirically with smokescreens and ambiguity, often including something of providing value to stakeholders and developing solutions to complex problems, but never detailing any specific tasks typical of a consultant.
In fact, there are many white collar professions about which that very question can be asked. You may be able to answer what a forensic investigator, actuary, or FP&A analyst does at work, but an investment banker, consultant, private equity manager, business analyst… these get trickier.
Lucky for us, however, the forum discussions prompted by Lukenas’s sudden death and Bank of America’s young analysts’ lofty desires of 100 hour work weeks brought insights from consultants and investment bankers themselves. And many of the specific tasks demanded of these investment bankers, consultants, and business analysts might surprise you including the follwing:
- Composing power points and pitch decks.
- Financial modeling and investment analysis.
- Legal research.
- Writing emails.
- Meetings upon meetings.
The employers creating these banking and consulting jobs are often multinationals that hold prestige and pay its employees handsomely. They attract graduates of top universities and some of the most talented, experienced, and competent professionals in the country. For this reason, it would make someone think twice before admitting that banking, consulting, and other professionals may soon become replaced with AI. However, it’s hard to look at the list above and not posit that AI will soon come after these consulting and banking professionals.
To make circumstances more dire, consider the commitments to AI that these financial and consulting firms have made:
- Goldman Sachs ‒ Marco Argenti, chief information officer for Goldman Sachs, told CNBC Finance in January 2025 that it recently launched a tailored AI model to a portion of its workforce, which “will eventually take on the traits of a seasoned Goldman employee.”
- Ernst & Young ‒ EY announced an investment of $1.4B into artificial intelligence in 2023. Since then, it has launched a tailored generative AI model to enhance performance among its 400,000 employees.
- JPMorgan ‒ In a February 2025 interview with Wall Street Journal, Teresa Heitsenrether, JPMorgan’s chief data and analytics officer, claimed the firm has roughly 100,000 of its employees using its tailored generative AI tool every day. When asked how it will affect the firm’s workforce, Teresa said the firm sees it “as a combination of humans and AI models, being able to do more.” JPMorgan intends to integrate proprietary firm data and give its generative AI tool more autonomy and reasoning capacity in the coming months, allowing it to make decisions and form strategies on its own.
Further, Bloomberg Intelligence reported earlier this year that global banks will see job cuts up to 200,000 over the next five years due to artificial intelligence replacing the tasks of human workers. And if bankers and consultants spend the majority of their time on the aforementioned list of tasks, it shouldn’t take much time for these tailored generative AI models to take a large chunk of composing slides, writing emails, and taking meeting notes.
However, the dissenting argument to Bloomberg’s report and to many economists and employment models throughout history is that new technology enhances employee output. Let’s say that a young Goldman banker costs the firm $300,000 per year and earns the firm $500,000 per year. This Goldman banker works 80-hour weeks and spends his time toggling pre-built financial models, formatting power point slides down to the millimeter, and spending roughly four hours a day in project meetings and calls. However, the banker also spends his time building client relationships at dinners and events, exploring strategies for its clients with firm partners, and researching how shifting financial data and new policies should be integrated into existing financial models. One might say the former tasks are an embarrassing waste of the country’s best and brightest talent, while the latter tasks are more fitting. Now let’s say that Goldman Sachs perfects its AI assistant to perform 80 percent of the more menial tasks—now it generates power points for the banker’s review, can give financial ratios and outputs for various input scenarios into financial models, and sends the banker a transcript and notes from meetings where his attendance was not essential. This type of AI would allow the Goldman banker to spend more time doing the latter tasks, work a sixty-hour work week, and perhaps earn the firm $600,000. In this case, Goldman would create new jobs for bankers, because the increased output justifies increasing employment.
Of course, this dissenting argument assumes (1) professional services firms will have confidence in its ability to continually increase sales, and (2) artificial intelligence will not push entire professional services industries to competitively lower its costs. These assumptions are increasingly difficult to make, as (1) many professional services firms charge hourly fees, and could charge clients less in total fees due to artificial intelligence models performing a significant portion of project tasks, and (2) consulting and M&A advisory services may see significant cuts due to tariffs, high costs of capital, and looming threats of recession reducing firms’ discretionary spending.
The most probable scenario is that artificial intelligence will make high profile professional services employees more profitable, but will also eliminate nearly half of their day-to-day work, leading to an even more competitive job market and yet even higher employee pay, especially as employers shed menial tasks from job descriptions. Thus, high profile professional services, such as private equity management, investment banking, and consulting will see an industry shift just as mining and manufacturing in previous decades in developed countries has experienced, with new technology and equipment demanding higher skilled workers who then produce more for the firm employing that technology and in turn are paid more than their lesser skilled counterparts.
As AI absorbs routine tasks, professional services firms will prioritize three irreplaceable human capabilities: AI oversight (quality control and model refinement), creative problem solving and automation with AI tools, and relationship-based sales. The shift should be a call for universities to fully integrate AI management and sales training into its MBA and finance coursework, and for economic development organizations to strengthen relationships with its metros’ office managers of Fortune 1000 companies in banking, consulting, and professional services sectors as these companies face threats of downsizing and job shifts in the near future. In addition to bracing for impact, communities can even take advantage of the shift with strategic AI partnerships between universities and professional services firms, funneling talented students directly into firms’ growing departments, or by introducing new public funding and incentives targeted to encourage job retention and growth among banking, consulting, and professional services firms.
Artificial intelligence is no different from technology as a whole in that it is far too often criticized as an “employee replacer” instead of an “employee enhancer.” Unemployment prediction models often account for activities AI can replace far more than the output that AI will add to employees’ work. While employees of professional services firms may blame artificial intelligence for an increasing lack of job security, they will hopefully soon be able to thank it for putting 100-hour work weeks full of formatting power points behind them.
