In light of Chat GPT 4-o what jobs in finance are at most risk of automation?
After seeing OpenAi’s announcement on Chat Gpt 4-o I’m completely amazed at how far AI has come in such a short amount of time. What is everyones consensus on the implications of AI on IB/ER/AM/PE/S&T etc. When will these companies start to adopt and engrain AI into day to day work? Curious to hear thoughts across the board and what jobs are at most risk of automation.
Based on the WSO Dataset, the finance roles most at risk of automation include those that involve repetitive and predictable processes. Here’s a breakdown of how various finance sectors might be impacted:
Capital Markets (ECM/DCM/LevFin): These roles, particularly in ECM and DCM, involve running predictable processes which makes them more susceptible to automation. LevFin, however, might retain more human elements due to its advisory component.
Mergers and Acquisitions (M&A): This area is less likely to be automated due to the need for negotiation and the variability in deal structures and processes. M&A requires a human touch for strategy discussions and negotiations, which are hard to replicate with AI.
Sales and Trading (S&T): The trend towards automation is strong here, especially in more standardized and liquid markets. However, roles dealing with more complex and illiquid products might still require human expertise.
Asset Management (AM): While some aspects of asset management, like portfolio rebalancing, can be automated, the strategic elements such as deciding investment theses based on evolving market conditions still require human insight.
Private Equity (PE): PE involves a significant amount of qualitative analysis, such as assessing management quality and market position, which are less prone to automation. However, some quantitative aspects might see increased automation.
Equity Research (ER): Automation can handle data collection and basic analysis, but the deep, nuanced understanding needed for high-quality equity research still relies heavily on human expertise.
In terms of adoption, many firms are already integrating AI to handle data-heavy tasks to increase efficiency and reduce errors. The timeline for broader AI integration varies by company and sector, influenced by regulatory considerations, technological advancements, and organizational culture. The consensus seems to be that while AI will automate certain tasks, it will also create opportunities for finance professionals to focus on higher-value activities that require human judgment and creativity.
Sources: Will robots replace your consulting or financial career?, Automation in fundamental finance roles, Future of Banking, Is S&T dying? COVID-19 impacts, Is Traditional Finance Dead?
bump
In my opinion I don't see a reason to pay any IB analyst 200k just so AI can do their jobs better and way faster. But I am aware I am a millennial that have grown up with tech and that's not who most of the MDs are. Sucks especially for the Gen Z I guess.
I doubt any banks will adopt it due to confidentiality issues.
Wouldn’t well-configured AI be more secure than fresh college grads?
It should be, but there is always the concern that the companies save the data for themselves. Hacking is the least of these bank's worries.
Hallucinations are still a massive problem. These models can get basic things wrong and are not reliable when you need 100% accuracy. They don't have a clear path to fixing this or they would've done it by now. Scaling parameters and increasing training data hasn't really helped, and there's a view that we're close to exhausting available training data.
As a VP/MD I'd much rather pay a smart analyst than trust one of these models in their current state, except maybe the most basic stuff like writing boilerplate text.
Excel Modeling & Power Point Slide creation will be automatic in the somewhat near future. You’ll just need someone to look it over to make sure it makes sense, which is what managers do anyhow.
No need for kids to spend 12+ hours a day doing it.
I already use Claude + ChatGPT to help me with analysis for CRE investments. Math might need to be double checked but sooner or later, AI modeling is going to be more commonplace.
Check out Clik.ai
What does Claude ai do? Doesn't Chat GPT do everything Claude can do?
Mid /back office are at the most danger.
But fyi, ibanking analysts and associates are also middle office strictly speaking, because they support senior bankers and are not bringing in deals / directly negotiating with clients. Some ibanking VPs are also middle office if their job is just execution, not deal sourcing. So all of them are at danger. Only the deal-bringing VP / MD and relationship hires (nepo kids) are safe.
Advice: climb to VP/MD asap, build client connections and focus on sourcing deals. The day-to-day grunt work is at the brink of being automated and the progress is very closely there and could happen any day in 2026 - 2028, or even before ethat
The back office is most in danger as the tasks there are the most rote. Front office tends to be more dynamic whether sales or analysis (or both) so should remain a net beneficiary for a long time
Though if your job is rote -- then I'd worry. IB analysts have incredibly rote tasks as do many associates. Only at VP level and above are you doing more of the dynamic work. PE associate is likewise a pretty rote task type of role, though not as much as the IB analyst / associate
I think with front office the common theme is that the jr. roles / entry level (up to 5yrs) is most at risk
Did you just learn the word rote?
Caveat that you're asking this on a finance forum where people's confirmatory bias and fears of losing their job/career will inadvertnly impact any response - so take the above (and this) with a pinch of salt.
AI is scalable intelligence. Bain are already shifting their model away from pyramid structure to more of a diamond (no juniors / more mid-senior people to interpret outputs of these things). This trend will only accelerate. View from many on the inside, and advising from the out, is with time a lot of the junior roles will be displaced and mid-senior roles become more efficient. VPs will have their own autonomous agents to run scenarios, decks, etc. without the need for comments and iteration as that can be fed into models which are trained on proprietary banking data.
My rather uninformed view is a parrot of one of the smarter people I know, which is the least likely to be disrupted for some time are the blue-collar roles. Plumbing, Call-out engineers, builders, etc. will be the last to be automated by robots and tech (c.10 years away until the prices have deflated enough for it to make sense). If I was starting my career again, i'd start there - but that's my 2c
I like the diamond shape analogy -- exactly what I was thinking in broad terms but that gets the point across very well. I couldn't agree more, AI will be a net pos for a long time for folks who are mid-senior levels. Junior classes will tremendously shrink over time
So what a college student do right now to stay competitive then if the job market is so gloomy at the junior level?
Build your network like crazy and hope you don't end up in a dead end job out of college. That's the brutal truth. Would be totally different if it was the late 2010s when I feel every college grad I knew was getting a way better job than many of my peers' first job after college in the early-mid 2010s. Can't predict economic cycles so best case is to be well prepared with a strong network.
One of my friends got their dream job this way at my MBA program. I don't think the FT role he got was even posted online. Did contract work for a startup and he was able to get a role with them after one of their recent money raises.
Google AI is currently telling people to eat glue because someone on Reddit posted that 12 or so years ago. I don’t think your analyst role is at risk in the next 2 years.
Change will come fast, but not that fast.
I don't disagree with the diamond structure, but how to you maintain that mid-senior level with a smaller junior class given already high levels of attrition at these places? Recruit more from industry? Fill with post-MBAs?
Are you serious- If you were starting your career again you'd be a plumber? Real rich coming from someone who sits behind a computer (myself included). "Well f*ck, these white collar jobs might get automated away. Wish I had learned plumbing so I could spend all day in people's basements and crawspaces, dealing with their piss and shit so I could make 90k a year."
Better to take your chances and still pursue white collar work than to just hang it up and resign yourself to the most risk adverse job you can possibly imagine.
I personally know a few plumbers and tradesmen who make far more than most bankers and PE guys on average, AND they are completely unassuming. The irony of it is that you and I probably have a better shot at making that work than making partner at a MF. And blah blah confirmations bias, whatever - how many LOADED MDs do you actually know? Me personally, not many. Pretty much all these guys are upper middle class by NYC standards and could not quit at their own discretion. Tied to their bonus and expensive lifestyles.
The trade guys that I know who have made it by working an apprenticeship for a few years to learn the job, then going out on their own and scaling their own businesses. Of course this will entail a lot of risk in starting your own company, however this is differentiated from some MBA dweeb trying to buy a small business because you can personally offer the skill set you are selling and can work solely for yourself if need be.
If were to do it over, I would probably go this route TBH. A big factor to also consider is that you wouldn’t have to work a bunch of banking LOSERS either.
Given 10 years and the will of a company's ownership (which is key), every single financial job that is not innately sales oriented is on the chopping block. For me specifically, my main hope is that I'm going to be high enough on the pyramid before the AI revolution to avoid the chopping block.
I think it will just make people more productive/work faster. Kind of like how spreadsheet software did.
Personally, I can produce high quality writing at ~2x the speed using AI. I just have it do things like help me when I get stuck, ideate, etc.
Quantitative stuff...it produces a lot of errors for now. But things are changing fast...
What do you do to be more productive / fast? Do you jot down ideas and have chatGPT write it to sound more eloquent or is there more to improve productivity?
My process is that I write an outline with the main points I want to cover, then ask GPT for anything I'm missing -> get more ideas/trim.
Then I write and if I get stuck I ask it to help me flesh things out, and usually remove ~95% what it writes, but that process gets me unstuck.
Enim quod excepturi est ducimus. Aut quia deserunt eligendi et. Quas doloremque inventore aliquid voluptatem rerum.
Explicabo deleniti impedit quos voluptate et quis similique ipsum. Dolorum consectetur voluptatem deserunt debitis magnam. Et deserunt consequatur unde.
Possimus accusamus tempora quaerat beatae omnis et minus. Optio officia a nisi est optio excepturi. Perferendis delectus quidem consequatur eius. Accusamus laboriosam quis architecto nostrum asperiores et. Eaque minus officia corporis nam cumque rem.
See All Comments - 100% Free
WSO depends on everyone being able to pitch in when they know something. Unlock with your email and get bonus: 6 financial modeling lessons free ($199 value)
or Unlock with your social account...
Fugit veritatis facilis nisi voluptates nulla quam. Et impedit qui accusamus temporibus eos voluptatibus veritatis. Et laboriosam nisi omnis inventore impedit ratione hic. Neque recusandae velit dolor delectus quo error tenetur.
Modi facilis quibusdam velit eos aut animi sequi. Ullam enim distinctio alias quaerat commodi repellat ex. Magni est enim eveniet possimus animi deleniti pariatur fugit.
Iure molestiae enim eos sint nobis dicta quis aspernatur. Et ratione est voluptatem ratione rerum dolorem. Incidunt sit laboriosam corporis officia fugit voluptas molestias dolores.
Dignissimos laboriosam sed impedit quo. Officiis facilis quas quos accusamus aut consequuntur provident dolorum. Nisi nisi quisquam dolorem. Rerum rerum amet quo et. Dicta sed totam possimus quibusdam ut harum suscipit.