The AI Memo That Moved Markets: Is a 2028 Economic Crisis Already in Motion?

That's what I worry that we're losing in venture it's either you're instantly 10 million in ARR two weeks in or like nobody cares. Well, not every good business follows that trajectory or looks that way. And this is sort of the cold water I've had to explain to some of our vendors in our portfolio. In order for like substance to matter, you have to have people who are looking for and investing in substance. Otherwise you're just yelling into the wind.

It's the week of February 23rd. Welcome to the Learning Corner, a Precursor Ventures podcast. Welcome to the Learning Corner by Precursor. I'm Mia Farnam and I'm joined by our GP and Managing partner Charles Hudson, where we will be talking through the articles and Goodreads that caught our team's attention over the past week. This week's first piece is the much circulated Citrini research piece the 2028 global intelligence crisis, written as a macro memo for the future, styled as a postmortem on how the AI productivity boom tipped into something much darker. A 38 drawdown in the S&P, unemployment at 10.2% and a consumer economy quietly hollowing out. The argument isn't that AI failed, it's that AI worked exactly as advertised. And that's the problem. The piece traces how agentic coding collapses the SaaS market, AI agents dismantle the entire intermediation layer of this economy and white collar job losses compound in a feedback loop with no natural break. It also digs into how PE backed software, LBOs and life insurance companies became a daisy chain of correlated bets on white collar productivity growth and what happens when those bets go bad. Last week we talked about something big is happening which rippled through group chats and LinkedIn and this Atrini 2028 memo seemed to have percolated all the way out to public markets with notable software sell off reflected in stock prices. Last week I know you asked yourself why this one speaking to why this article was picked over others and got the reaction that it did. Do you have the same reaction here?

I read it and I'm always like these posts are fun to read because they're like thought experiments. I've also learned there's very little I read in there where I'm like, I haven't heard other people make these comments and this is like the second time in two or three weeks where something has blown up and I was like oh, that's right, you live in the world of private markets where you read a bunch of people who've been Talking about and thinking about these things for a while and I have no idea idea what makes any given post break through and stand out, but it does. I read it and there were certain parts of it that I was like, I'm not entirely sure I understand. They don't really present like an economic model for how this all works. It's more of a thought experiment. There are two things I read in it that I thought were interesting. One is this whole notion that no one's going to buy software anymore. I think a lot of people are like, oh, people are going to vibe code stuff. Like there are other alternative universes where software is less valuable. And I don't know, I told our LPs this a few years ago that like, I think AI is going to be very deflationary for software. And they look to me like a crazy person. I think it's far less likely that you're going to vibe code an H R I S system. It's a lot to maintain. And I think the thing people seem to forget is if humans are going to vibe code all these things, everybody's going to have to recreate the same feature themselves and maintain this stuff and do patches and security and. And the history of software is it's more fun to build prototypes than it is to maintain production software. But that's like the human world. What I could imagine is something that looks more like mo book where you have a bunch of agents who get together and say we're going to build and maintain an HR system and we're going to build it together as autonomous agents and as our human overlords, or underlords as the case may be, encounter problems, we will collaboratively. But you know what that sounds like? That sounds like open source software. The only difference being the contributors are agents and not human beings. And if you think about some of the things that have plagued open source software, it's that people don't generally get paid to maintain it. And they have jobs and unless the project achieves gate velocity or has a really motivated project creator, they don't go anywhere. But the agents don't need jobs. So I can imagine a world where there is more software that's commonly used. It's supported by an army of agents and those agents are using some platform that multipook or some agent language where they are doing the development advancement of features. I can envision that world. I don't think that world happens in 2028. I think the piece also is your classic. We won't need McKinsey we won't need all, we won't need doordash. I was like, you know, coordination is of what intermediaries provide. And I read the DoorDash piece and I guess it could be like agents finding work for people, but this seems like decentralization for decentralization's sake. And I kept reading it and I was like the thing that bothered me about the piece, that I couldn't put my finger on it until this morning, is it assumes that there's no system level reactions as these things are happening. It's just like these things happen and people just allow it to happen and they don't intervene. And I don't think that's what's going to happen.

I mean we're already seeing people have pushback as data centers kind of take over in different rural areas. People get a little bit more personally attached to kind of where things are headed. I'm gonna fudge the author's name but run another piece that was a response to the article and you said the response of the software sell off. People are kind of forgetting that a lot of these software tools in stronger language stink to use day to day. So it's no surprise that there's pushback. The market is reflecting so a portion obviously to do with different launches of products coming out from foundational LLMs, but maybe more to do with the fact that people are always wanting more improvements in the multi million dollar contracts that they're paying for and for the first time maybe those folks are looking at other options. But we talked about this two podcasts ago. The idea that everyone moves over immediately, massive Fortune 500 companies move over immediately and abandon these software companies that have serviced them for the last two decades across their time organization overnight. It just feels a little bit reaching. Noah Smith wrote a really good piece I thought about in the public's reaction to the piece on his substack no opinion noting that it's normal for stocks to dip when a new product launches under Foundational LLM. But the fact that this was 1 like on a substack and 2 just a scenario and still made so many headlines was what surprised him. He also to your point about we've seen this before. He draws parallel to 2008 where different looking stakeholders sold each other assets that simply weren't worth what they were priced and the market collapsed under the volume. So any thoughts on that comparison from your perspective?

Every time I talk to somebody who's in credit, the only thing I want to talk to them is but it's Data centers, I think the only thing I want to talk about, it's also the only thing they're investing in. So it's actually a pretty easy conversation. And I think if there were a canary in the coal mine, I do think it's. I think we mentioned this before, when these things blow up, it's usually on the debt and credit side of the ledger. It's not usually on the equity side. We don't know the residual value of a data center, particularly on the current trajectory, because most data centers are power constrained. And so right now, two things. One, there appears to be relatively little. It's like this weird conundrum. We don't have enough data center compute to power everything that people want to build. But at the same time, it's not clear that the ROI on building all of these data centers, everyone's like, it's going to be like fiber. We eventually needed it. I'm like, well, we put it in the ground. It wasn't a big eyesore, it wasn't power constrained and we lit it up as we needed. Data centers are a little different. I could tell you a world, if chips remain power hungry, where some of these old data centers simply just don't have enough power to be relevant. And I think this is the first, I think also, as you mentioned, the social political piece of undefeated. This is the first tech wave in my life where people were not unambiguously enthusiastic about it. And I don't think people are wrong to be hesitant in the like, what does this mean for me, Bucket? And I do think AI is going to reshape some political alliances in terms of like who's for it and who's against it. So I don't know. I think a lot about all of the credit bets are being placed mostly on data centers and that scares me a little bit.

And again, similar to what Will talked about in tool shaped objects last week that we featured, the piece did exactly what it was designed to do, which was to get everyone talking about, where do I fall in this chain? The answer was everyone is impacted and it eventually will come for you. I mean, they walk through kind of like the drawdown on consumer spend. Eventually, like no one will have enough money to be able to buy anything. So even if you sell a CPG product that no one will want or a luxury good that's made by hand, like, you will be affected as well. Again, if I lived outside of this bubble, which you and I live in, which is just venture and tech news all the time, you would know that there's been hundreds of these pieces published prior and a lot of them kind of touch on similar threads, but if this was the first time that I'd read it but it reminded me of like Tim Urban's Wait, but why? That he did in like 2015 or something and everyone rediscovered it. Similar reaction, everyone running around like crazy on Reddit. I don't think the full Citrine scenario is where things end up, but specific points along that plane do feel entirely possible. But as early stage investors actively funding in technology that could accelerate parts of this, do we have an obligation to think about what we fund and what we don't? I know that there are specific buckets that this definitely impacts that we don't place enough capital in to move the needle as heavily in credit, but did you have any thoughts on that?

I talk to our LPs a lot. My view is investing is like voting for a version of the world that you want to see exist. So every time we make an investment I always ask myself, do I want this version of the world to exist? And there are some very relevant, interesting technology areas. I'm far less interested in prediction markets than a lot of my friends. I think it's mostly speculation and gambling. And does it sometimes reveal interesting information? Yes. I'm just not interested in that version of the world. And I don't say that to judge people who are. I'm just like, that's not something I want to usher in, so I'm not going to put my money behind it and it doesn't matter. Pure point. There's plenty of people who do want to see that version of the world exist and I'll let them fund that stuff. But I do think you can't have it both ways. You can't say I don't care about the social impact or it's not important. And then when it happens, be like, I'm shocked. I'm shocked that there's gambling. You can't have it both ways. So I I just find it's easier to just say, look, you're putting people in business to achieve their vision for the future. Just make sure that you're interested in the version of the future world that they're describing.

Well, for our next topic, we're highlighting a great piece from Strictly VC Download where Connie Loises sat down with Stacey Brown Philpott, former CEO of TaskRabbit and the founder of Our Friends over at Cherry Rock Capital, a firm laser on series investing. Cherry Rock is deliberately focused on Series A, a Stage Stacy believes is full of opportunity, with her team's edge being the ability to find great companies early across multiple coasts and position themselves as the obvious partner before the deal gets super competitive. Her approach to winning in that competitive landscape is finding companies early and positioning Cherry Rock to be the obvious partner by the time the company is showing real signs of growth. She is focused on quality revenue over arrangement as the signal, specifically long term contracts and deep enterprise relationships that point to product market fit at scale. The episode also touches on the bifurcation happening across venture as fund sizes grow with lots of activity at seed and at the mega rounds but a real scarcity of co investors and follow on capital at series A and B. Stacey's view is quite simple. It's great businesses will get funded. She also shares a measure take on workforce displacement, noting that most founders she speaks with are focused on workforce enhancement and that she believes we are still years away from true displacement being a real concern. Stacy's framing of quality revenue over ARR as the signal for series A traction is something we've talked a lot about before. When we've talked about like experimental ARR, it seems like years ago when we used that what can we do to position our founders to lead with real metrics even when it can be tempting to lean on something a bit more glamorous but fleeting? I find myself having conversations with our founders all the time of like where can we strike the balance of making sure that you're competitive enough, knowing the types of more glamorous metrics that are getting tossed out there, but also being true to the true traction that you have and not over embellishing.

I listened to this and I thought about two things. One is as we have this great pulling apart of venture between big firms and little firms, the question is well what does it mean to be in the middle? And I think Stacy's got some good arguments and I do think there are going to be companies that are not immediate vertical rocket ships that either need more time or need more capital or need a partner who can like help them get from A to B. I think that's what I worry that we're losing in venture. It's either you're instantly 10 million in ARR two weeks in or nobody cares. Well, not every good business follows that trajectory or looks that way. And this is sort of the cold water I've had to explain to some of our vendors in our portfolio. In order for substance to matter you have to have people who are looking for and investing in substance. Otherwise you're just like yelling into the wind. I look at our portfolio, we have a handful of companies that are compounding really nicely. They're, they're not growing 10x. This is real revenue. These are customers who are using and paying for the product and coming back month over month. And I have, I have no concern. There are other companies that I've met where I've seen vertical growth charts and I'm just like, are those people ever going to come back and use this again? And we see this at the beginning of every cycle. You know, I think back to mobile and for a while it was downloads. Everyone's like, look at this downloads number. And then we eventually realized, you know, what's more important than downloads? Usage. You know, what's better than monthly usage, daily usage? You know, it's really interesting, the ratio of debt. And eventually we figured out what mattered. And before we figured out what mattered, a lot of people blew a lot of money on things that ended up not mattering. I'm certain we were doing the same thing in AI right now. And we will look back in a few years and say, wow, some of that stuff was substantive and real and some of that stuff was selling a dollar of compute for 75 cents. And it shouldn't be shocking that if you're giving away money, people will be the path to your door.

I love the Cherry Rock team. I know that they do a really good job of identifying people who are kind of like at the inflection point and getting ahead of them before, you know, it hits the automatic series A metrics which everyone then has already flooded to the market and typically given term sheets that not everyone can compete on. But that strategy does require everyone around the investment committee table to be bought in on being okay with a little bit less traction and metrics than you'd be comfortable with if you're like a true series A or true series B investor. I think this a lot about. I was on the phone with somebody else who they are are a classic series A fund and you can tell they want to come earlier and earlier. Specifically like with pre seed stuff. They were asking me like what sort of things could be competitive for us to construct within our own fund model to make us take up a bigger space within that market. And my response was does your leadership fully align with what pre seat looks like? Because if not then you're just going to end up getting into deals that aren't actually the best pre seat. They just look good for what you're trying to do. We do pre product, pre everything. And so we're very comfortable with that. But I know a lot of people who attempt to move early as a mechanism to get competitive end up like almost selecting the wrong types of deals

because you get the worst. Yeah.

So how do you. I don't know. I'm curious to know what you think about that when we talk to you or thinking similar strategies.

I think it's funny. VCS make fun of people for being Luddites. I think most people in venture don't want the venture business to change just as much as people in the newspaper business probably didn't want the newspaper business to go digital. And I think the most dangerous thing you can do as a fund manager is assume that the future will be like the past. And I think there was a very important and clear role for 500 million doll series A venture funds in a world where fundraising was sequential and where the biggest venture funds weren't actually that big. We just don't live in that world anymore. And so the question's like, if you're going to participate not at pre seed or seed and not at scaled Series A, you got to have a reason for existing and one reason for existing. If I were to like play back something I think Stacy said kind of without saying it, is there are companies that would have been Series A ready three years ago based on metrics and progress that are good companies that are being overlooked because there's a set of investors that are looking for something that's an order of value higher. And the mistake I see people make is what you describe, which is like, we want to do pre seed. Someone's like, yeah, we do pre seed companies. I'm like, what are you looking for? Well, companies with at least 250k in ARR. That's what our seed portfolio looks like, not our pre seed. So you might find companies with those metrics, but they're probably not the droids you're looking for. Like, they're not going to be the best companies.

Well, our last piece this week is a LinkedIn post from Alex Paul, one half of the chainsmokers and founder of Mantis vc. Alex shares that over the last few weeks he's received a handful of eerily similar emails from portfolio companies, founding teams splitting up and moving on, not because the companies failed, but because the frontier had moved. These are mostly 2021 vintage companies that tried to adapt AI, some really thoughtfully and some by just bolting it on. And now the founders are staring at a brutal opportunity cost question, watching entirely new categories get created in real time. With teams shipping faster than ever and ARR curves going parabolic, optimizing an older business no longer feels like the best use of their time and ambition. Alex frames it not as a failure, but as a first principles moment. Sometimes the right move is admitting you gave it a real shot. We learned a lot and now it's time to go build the thing you actually want to wake up thinking about again. Charles, how often are you seeing this type of behavior across our portfolio or with founders that you know personally? And for those conversations, how do you make the decision between encouraging the pivot or abandonment altogether or encouraging them to see their original idea? Especially with the market that we exist in today.

I just think as an investor, you can't make somebody keep going. So usually if someone tells me like, oh, you want to quit? I'm like, well, it sort of doesn't matter why unless I think they're just tired and they need a break. I'm like, okay, you've decided to quit, as is your right to do so. And so I think there's something even more I don't say probably more interesting that I've seen lately, mia, which is I've seen brand new companies breaking up quickly, like six months in, and I've talked to a lot of them. And in many cases the pace of AI development is such that the founding principles for the company, a new model release comes out, a new technology development happens, and these folks are instantly like, oh, the whole reason we started this company is now in peril. And the question is like, what do we do? And in many cases there is not agreement on the answer to that question. One person's like, I think we should go left. The other person's like, I think we should go right. In response to this development, I'm seeing a lot more team breakups. But it used to be that the team breakups were more interpersonal conflict or like significant differences in strategy. And now I guess what I'm seeing more of is variable perspective on how to respond to the changes in the AI ecosystem and a lack of agreement about both the optimal response and the severity of the threat of something. And in many cases, like, we just had a brand new team that recently just announced a seed round that's maybe a nine month old company. Co founders are parting ways largely over a disagreement in sort of what all these AI developments mean for their strategy.

So that's fascinating. Alex is usually talking about this in the terms of like successful series A or series B companies, but we're Also seeing this happen right away, just with the current market. Last question here. There's something in this post about the relationship between excitement and output that I find quite interesting. Founders having fun building again actually produces perhaps meaningfully better companies. Do you believe that's true? And if so, what does it mean for you on how you evaluate founding teams going forward? I'm honestly, I'm thinking about conversations that I've had with almost a lot of, like, repeat founders who are so pumped to get back into this market and, like, build again quicker and leaner. And then sometimes when I hear of their ideas, I'm kind of surprised with what they've chosen to spend their time on.

It's so funny. Yesterday I was talking to one of our portfolio founders. This is someone who I am now working with for the third time, and we were talking, and he was showing me a demo. I was like, you know, I've never worried about speed and velocity with you. That's never been an issue. And he kind of laughed. And he goes, we use our own product every day. I go to him and he goes, well, it's hard not to feel urgency when you're using your own product every day and you see something wrong. The desire to fix it is really strong. And he's like, I don't know how you can have the same level of urgency for a product where you don't have a connection either to the customer or to the problem. And I was like. Because I was like, I wish I could teach people to be more urgent. He's like, you probably can. I'm like, I know. And he's like, well, we. We simply care because we are part of the audience. And I think we talk a lot about how that can blind you as a founder. I think it can also be incredibly motivating. And so we've had a few founders where sometimes I'll ask them, like, do you even care about this problem anymore? I don't want to be rude. And sometimes they're like, no. But, like, I signed up to help these poor tax accountants do ledger reconciliation, and I can't fake my enthusiasm anymore. And so my hope is that everybody gets to work on something that they're excited about. It doesn't have to be because it's like a glitzy, glamorous thing, but it's because it's meaningful to them that does seem to produce better outcomes.

Well, we are all out of time today. Thank you, Charles, for jumping in. I know you're traveling. Thank you to our listeners for hopping in as well, Please like and subscribe. And we'll talk with you all next week. Thank you. Bye, Sam.