PE Associate

I posted the other day that senior leaders in finance have already made their money and don’t care enough anymore to still be in the detail. Here is the perfect example. Hero to many (including myself) uncritically trusting the results of a probabilistic LLM. I see this a lot at my fund as well, sad to see

@kylekuzma @litcapital

@kylekuzma This is the craziest thing I’ve ever seen in my life

Yep wait till their data center build out 🔥🔥🔥 heavy capex

@nic_carter I just want to say that there is no AI model today that can create sophisticated LBO models. I can’t challenge the other pieces because I’m not experts on those things but complex LBO models they cannot do

The “it’s not AGI because machine intelligence is jagged” is dumb cope. It’s obviously AGI. If you had a friend who had a 130 IQ, could write production code flawlessly, could write academic papers of a high research caliber, pass any exam in any field with flying colors, create a sophisticate LBO model, draw technical diagrams perfectly, compose poetry in any language, and could find solutions to significant unsolved mathematical problems, you would call that person a world historical genius. Certainly, no single human has ever had intelligence that “general” before. Now you think it’s “not AGI” because it sometimes slips up and makes mistakes - so does any human that you would consider “extraordinarily intelligent.” The professor might forget a colleagues name that he has known for a decade. He is still considered intelligent. The math genius might be a little autistic and shy, unable to maintain polite conversation. Still intelligent. You might stare at the fridge for 30 seconds unable to find the butter, despite 5 million years of evolution perfecting your visual intelligence. We give intelligent humans a pass when they have jagged intelligence. So why the double standard? The qualities people list as “necessary for AGI” are important traits to have, but no longer pertain to intelligence. People will say things like “true AGI requires agency, long term goal setting, embodiment, self-direct action”. But none of those things are intelligence. Those are “things that humans have that AI lacks”. Raw intelligence, AI has it in spades. That other stuff - important yet, but broader than and different from intelligence. The unwillingness of people to acknowledge that AGI obviously exists and has existed for a while is due to a kind of anthropic chauvinism - a psychological need to believe that humans are superior in every respect, that we possess soft skills that no machine could replicate. Yes humans are different from machines, but if we are limiting the discussion solely to general intelligence, AI has it already. That battle is over. If you want to reframe the discussion to matters of human dignity and personhood, fine, but that’s not an AGI question. That’s something else. Just take the loss on AGI already. It’s over.

@B3nL1pman @spitspinner @BoringBiz_ How can he lose money when his cost base is so low

@spitspinner @BoringBiz_ Yeah I’m sure Elon didn’t think this through and he’s going to lose $ lol

SpaceX IPO valuation implies a 93x revenue multiple and you don’t even have a P/E ratio because the company has negative earnings

+ adding to this, a fed hike will also bring the show to a halt. I’m not sure what they can do against this inflation backdrop. We’ve already learned that inflation is not transitory…

Calling it now so it’s on the record for this account. I cannot see any scenario where we aren’t on track for a severe market correction. Absolutely peak NTM earnings multiples layered on top of peak earnings which are structurally non-recurring and financed by primary equity / debt (rather than an organic earnings base). Not sure what the catalyst will be (maybe systemic risk in private credit or maybe Iran) but I really believe we are either at or close to the top of an equity market that will implode.

@sxhealth101 Actually those are just investment banking analysts you are seeing, think you have gotten the two mixed up

I can tell a p0rn and masturbation addict when I see one. They always have the same visible characteristics: 1.He has a “dead” look in his eyes, the eyes look empty and lifeless, like life has been sucked out of him.

@clay_phi @haider1 How?

@PE_Associate @haider1 And yet he’s still wrong.

Yann LeCun says you cannot build a reliable agentic system without a world model LLMs don't have world models. They can't predict the consequences of their actions before taking them "they just act, and whatever happens next is someone else's problem" Without that, it's not intelligence

@clay_phi @haider1 “This guy” is the modern forefather of LLMs…basically everything you see today is built on top of his research (which he won a Turing prize for)

@haider1 Um…an LLM is a….probabilistic model….literally the opposite of what this guy is saying

@staysaasy Was going to say, even without AI those roles are all getting shifted offshore

Ok let’s go to the data from indeed hiring lab (where this data came from) Graph one: oh sick, customer service jobs postings are up Graph two: oh actually when you filter for new job posts it shows it’s clearly still declining, like it has for years now, and is the lowest since the start of the graph. Clearly this means the uptick in the other graph means jobs are simply not getting filled and are being left up on the site. Literally all of twitter: LOOK CS JOBS ARE UP YOU IDIOTS!

@mil000 That’s not how prenups work. You carve out assets (i.e. my stake in Kalshi) not $ value pre cap gain

she timed this so it would be right after the raise ie in divorce she wouldn’t split as she went in with the wealth

We run a rolling internship program and we genuinely had an intern come in and use the Claude Excel plugin on the live version of a live deal model (they were just asked to add some data to a data sheet at the back). It deleted the data sheet driving the model to create a new one (which cannot be undone) so the team had to go back a version…

I’m extremely interested in what ppl think of 2026 interns in a couple months.

The key word here is “potential”. It is so easy to pilot a product and conclude there are massive potential efficiency gains based on a user survey. It is so difficult to ever achieve those gains. There isn’t a single PE portco that I’ve heard of where AI is driving real improvement

Just spoke with a couple friends who work on the operational side at very large private equity firms Every PE firm is now scrambling to find and recruit AI talent who can implement the newest tools into their PortCos This includes former and current technology executives who are familiar with the industry, as well as engineers who can drop directly into the business to build custom made tools Will keep the comp numbers confidential but they are absolutely insane, especially at the senior levels For the PortCos that have already piloted this, the efficiency gains are huge. Some are well within the range of 20%+ headcount reduction potential for back office functions

@BlueyAnon @litcapital 😂😂😂 all time

This is brutal to watch lmao

@ar0cket1 @giffmana Can you explain your view a bit more? I thought the original point that this is evidence that RL transfers / generalises was sensible so keen to understand why you think it’s so obviously wrong

@giffmana just because it leaks outside of the prompt isn't generalization but just inherent to what RL does. its the same reason that non reasoning models of models start outputting think blocks and start thinking

In other words: their RL transfers/generalizes.

@justintime277 @giffmana @BoringBiz_ Google gets paid on a per click basis and Google search revenue is up 19%…

@PE_Associate @giffmana @BoringBiz_ less clicks to websites from queries = less time on the websites = more time on google instead = more queries when spending more time on google

How is Google search still growing 19% year over year with nearly monopoly like market share in the category? Legitimately might be the greatest business ever created in the history of capitalism

This doesn’t explain why revenue would go up though. Google ads run on fixed daily budgets at the customer level. So for revenue to go up by 19% (lower on a CC basis), total customer budget would have to increase by 19% (which can really only happen if Google is delivering double digit % more end market sales through its ads). It isn’t obvious to me that any form of current AI can explain that

@PE_Associate @giffmana @BoringBiz_ I guess its that When u use an LLM, the LLM use web search through google.

@giffmana @BoringBiz_ It isn’t really obvious to me. Grateful if you can explain your view

@BoringBiz_ You read what he wrote, right? > Search queries are at an all-time high with AI continuing to drive usage I think if you think about this for a few more seconds it's clear how search can still grow? It just doesn't really imply the same thing it did 5y ago though.

@jimcramer Oh ffs I just told my mum to go long intel. Now she’s going to be broke. Thanks Jim

Intel is such a horse. I have NO bear case

Man I love your content but surely you acknowledge that the model presented is below the standard/complexity of even a private equity interview case study. I’m sure these models will get to a stage where they can provide value add on excel modelling for actual deals but none of the current solutions (Shortcut, Mosaic, even Claude in Excel) can handle the complexity of a live deal model. Have spoken to 30+ PE associates across a range of firms on this and it is a genuine consensus view

I’ve been playing around with @tryshortcutai to see what I can have AI build for me and the end product is wild. I wanted to stress test Shortcut into building a master excel file for everything I would need for an investment committee. I had it build: 1) A 3-statement model 2) Toggles for multiple cases 3) Charts and tables for IC Memo must haves - revenue by segment, customer concentration, etc. 4) EBITDA Bridge charts 5) A Capital Structure chart You can pretty much have Shortcut model for you while you go and focus on the written parts of the IC Memo. Lmk if you’re interested in learning more, the key here is once you’re a wizard at prompting, the models are incredible.