New agent released yesterday scored a 41% on PowerPoint, Excel modelling, etc. vs. junior analyst level work. Now the question is will this lead to better hours or less analysts? GPT came out about 2.5 years ago so it's safe to assume that 100% will be hit relatively soon (probably within the next 1-2 years).
| +404 | Evercore Intern Seizure | 62 | 59m |
| +98 | UBS IB Americas has failed, now behind Santander and Stiffel | 27 | 1h |
| +79 | JPM M&A is Gone??? Purely Coverage Banking??? | 41 | 8m |
| +60 | How do I understand vs. just memorizing? | 11 | 4h |
| +48 | Losing my personality in Banking | 9 | 1d |
| +45 | The good and bad with Wells Fargo | 12 | 3h |
| +42 | Associate & Above IB exits | 18 | 6h |
| +35 | Sent my Claude prompt to 200+ Teams chat. MD wants to see me Monday. | 18 | 8h |
| +33 | Incoming IB Analyst: Best Ways to Prepare? | 13 | 3h |
| +31 | Tech to IB Pivot | 20 | 17h |
Career Resources
Get instant access to lessons taught by experienced private equity pros and bulge bracket investment bankers including financial statement modeling, DCF, M&A, LBO, Comps and Excel Modeling.
less analysts and higher workload, you'll be spending the time pumping out more then fine tuning them when things are bad/hallucinated. Gonna be a lot of documents half-baked by an MD where all the formulas are absolute garbage and you'll have to write it from scratch
Can I see the link to source
17:36
thanks
I still believe manual labor would be better than using an AI model or any LLM model. I say this not because sure jobs might be more competitive than they already are for banking or because there will be less openings for banking but because these models can’t pick up the phone and talk to clients and walk them through what’s going on. Sure, there’s probably a MD who can do that but they probably have so much other things on hand that they wouldn’t have time to run the ops for this on their own. On top of that, banking requires a lot of personality and human like behavior imo… and an AI or generative model wouldn’t be able to act like so. Apart from that, I feel like manual labor is still more effective in terms of catching small or minor mistakes and gpt can be used as a verifier to catch anything small as well but not to build out a full blown out comprehensive model that can be inaccurate. It is only trained on past data, not current data, so cannot create projected valuations based on current market expectations unless the data is being constantly fed.
Architecto eaque reiciendis et ut quia est itaque. Dolores odio eos quaerat voluptatem expedita rem.
Nesciunt quisquam doloremque laborum provident. Ea sed est quod velit ad qui. Officiis explicabo iure assumenda nostrum nihil.
Similique in consequatur necessitatibus quae sequi dignissimos magnam doloribus. Et est voluptas aut sed in. Illo autem impedit temporibus optio cum eum dolor hic.
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...