Pod Insights a16z Podcast How AI is Reshaping Labor Markets: A $Trillion-Dollar Opportunity Explained About this episode Did you know the U. S. nurse labor market is over $600 billion annually, but the dedicated software market for nurses is almost zero?
In this episode, General Partners Alex Rampell, David Haber, and Angela Strange discuss how AI is revolutionizing labor by automating tasks traditionally done by humans.
They’ll trace the evolution of cloud eras — from the original to financial services-enabled to the current AI-enabled outcomes era — showcasing how AI is creating unprecedented opportunities, allowing startups to outpace incumbents. They also explore how this shift will reshape industries, where we are in the adoption curve and what companies need to succeed, and the gaps where the a16z Enterprise team would love to see more innovation.
Timecodes:
00:00: Introduction
00:42: Historical Evolution of Labor and Technology
01:58: The Digitization of Filing Cabinets
04:32: The Cloud Revolution
06:43: AI and the Future of Labor
07:39: Economic Implications of AI in the Workforce
12:29: The Role of Systems of Record
17:37: Challenges and Opportunities for Incumbent Software Companies
20:55: The Messy Inbox Problem
22:01: AI's Role in Transforming Workflows
22:31: Case Study: Tennr Healthcare
23:36: Defensibility and Differentiation in AI
26:01: Emerging Opportunities in Compliance
29:26: AI's Impact on Job Roles and Market Dynamics
33:11: Metrics and Market Size
36:47: AI in Professional Services and Legal
42:26: Future Directions and Opportunities in AI
Resources:
AI Turns Capital to Labor: https://a16z.com/ai-turns-capital-to-labor/
The Messy Inbox Problem:
https://a16z.com/the-messy-inbox-problem-ai-apps-wedge-strategies/
Vertical SaaS: Now with AI Inside: https://a16z.com/vertical-saas-now-with-ai-inside/
BarbAIrians at the Gate: The Financial Opportunity of AI: https://a16z.com/financial-opportunity-of-ai/
Stay Updated:
Let us know what you think: https://ratethispodcast.com/a16z
Find a16z on Twitter: https://twitter.com/a16z
Find a16z on LinkedIn: https://www.linkedin.com/company/a16z
Subscribe on your favorite podcast app: https://a16z.simplecast.com/
Follow our host: https://twitter.com/stephsmithio
Please note that the content here is for informational purposes only; should NOT be taken as legal, business, tax, or investment advice or be used to evaluate any investment or security; and is not directed at any investors or potential investors in any a16z fund. a16z and its affiliates may maintain investments in the companies discussed. For more details please see a16z.com/disclosures.
Transcript 0:00 Speaker 0 Now you have software agents that are effectively doing what for 65 years had been human work. You
0:07 Speaker 2 know, is that going to increase software revenue 2x? It could potentially increase it 10x.
0:11 Speaker 0 It's not even on the same kind of playing field. So all the data is here in the cloud, all the compute is in the cloud, and now you just kind of mix them together. Many of these incumbents aren't gonna evolve.
0:20 Speaker 2 Every time there's a new technology shift, we have to challenge every investment thesis that we thought was not gonna work. And so many of them now are going to work.
0:28 Speaker 1 Motes still matter. And a lot of the motes in software today are the same that they've always been.
0:32 Speaker 0 Well, it's it's kind of both defense and offense for these companies to figure out what the hell to do.
0:41 Speaker 3 Alex, you wrote an article recently, input coffee, output code. But this idea of turning capital into labor, hasn't this always been true? What's new here?
0:51 Speaker 0 Well, it certainly has been true for a long time. If you watch some old movie about the Romans, you'd have all these Roman slave laborers or Roman uh soldiers that were rowing in unison on a boat, and then of course you got the steamship, and then you don't need these 50 people rowing anymore. Um, so clearly there has been this long period of time, this long historical arc of technology is in some cases augmenting labor. But it was always kind of the brawn and not the brains, right? It's like uh I have a bunch of people that are sewing clothes and now I have the loom.
1:25 Speaker 2 Right.
1:25 Speaker 0 But everything that was kind of what we would now call white collar, um, that hasn't happened before. And and what I talked about was uh the three or four different eras of of software, just story information. So for the longest time, if I wanted to keep track of who was on my airplane from the Wright brothers onward, um, I would have a filing cabinet with it's like, oh, here's Pan Am Flight 192, and here's who's on it, and I'm writing down their name, and then you know, they might call or send a telegram back in the old days of I don't want to be on that planet anymore, then I erase it and then I refile it. And one of the first examples of digitization of the filing cabinet was something called Saber. And this was developed by American Airlines, I think in 1959 or 1960 in concert with IBM. And this was one of the first examples of taking this filing cabinet that really kept track of who's on Pan Am Flight or American Airlines and putting it in a database.
2:19 Speaker 0 So now instead of having filing cabinets with lots of erasers and whiteout, you replaced the filing cabinet with a computer. And this in turn begot travel agents and travel agencies because now what was a travel agency and what was a travel agent. They had a thin client, a little computer, a terminal that I remember like booking airline tickets with my mom in the 1980s. It's like you'd go into the travel agent, and they had a like a green screen computer that connected to a mainframe in Texas, which is where Saber was. So the first realm of software was really kind of called from 1960 onwards when computers became a thing of take a filing cabinet. So it could be the HR filing cabinet. It could be the medical filing cabinet. It could be the financial filing cabinet and put that in software. So, and what was that?
3:05 Speaker 0 It's a database with a front end to actually go enter things. So Quicken famously did this in the 1980s for financial statements. There's a company called PeopleSoft that famously did this. It was the first HR filing cabinet, but put as software. Um, Saber obviously did this for airline tickets. Um email, I would even argue did it for mail. True, right? It's like you have files of mail, and now you just have uh a filing cabin if you want your computer. But the the actions that were done on the software were really the same. So imagine an HR department that has 50 people working in HR. They might have one person that's in charge of the filing cabinets, and the HR person says, hey, get me, get me David's file because I want to talk to him about something, which is very scary.
3:49 Speaker 1 Uh
3:49 Speaker 0 don't worry. Um, but get me David's file. So the the filing cabinet person, the gopher, this is what the filing cabinet person would often be called. Go for something. That's what it means. Go for there's the gopher that obviously Bill Murray tries to kill in Caddyshack, but then there's the gopher of the person who goes for something, right? It's like if you are a gopher at Creative Artist Agency, you were going and getting files. So that person went away. The filing cabinet went away. It was more efficient from a space perspective. But the 50 HR people are still 50 HR people today. So round one of software was really take the filing cabinet and put it um not as physical files, but as as a database with the front end. Round two, which started arguably at like 1998, 1999, and this is what Salesforce did.
4:34 Speaker 0 Like the idea of a customer relationship management product has been around for a long time. Or the Rolodex is an actual physical thing where you put every business card, organize them alphabetically, and you'd find the person that you want to contact that way. Salesforce put that not in software, but in the cloud. So what uh QuickBooks had done for a long time, or uh something called Great Plains software had done for a long time, NetSuite did in the cloud.
5:00 Speaker 3 Right.
5:01 Speaker 0 Zendesk put email support in the cloud. So it was still software, but instead of having a giant mainframe in your computer and in your office somewhere, you now had it in the cloud. It's just much, much easier. You don't have to have like a dedicated IT team to go like worry about your server like exploding in flames if your your office building burns down or something. So it's it's more secure. So everything that was kind of software 1.0 then became software 2.0, which is in the cloud. Um and that played out from call it 1998 to maybe 2010. And then that grew a little bit with the insertion of financial services, because the way that I like to think about this is how many restaurants need and will pay tens of thousands of dollars a year for software.
5:49 Speaker 0 Like Pan Am needed this in 1960, but does a restaurant with one location, are they going to spend 100,000 on a server and pay for software? Like, no, in 1960, no in 2000. But when the idea of bundling and payment processing became a thing in other financial services, now the market became big enough for restaurant software to exist because, and this is where toast came from. Toast is a 15 billion dollar company that does this, or service Titan, again, what's the software market in 1965 for HVAC contractor zero? What's the cloud software market for HVAC contractor zero? But once you bundle in these other things, it becomes big enough. So um, but the point that I'm getting to is that the same 50 HR people that worked in 1960 are the same 50 HR people in 2024.
6:38 Speaker 0 The same email support team in 2024 was the phone support bank in 1985, was the letter writing, you know, typewriter support team in 1965. What's exciting about AI is that it's taking this filing cabinet and now allowing actions on the filing cabinet. And that's what I think is really revolutionary, because you can actually ask the software filing cabinet application, like workday, hey, I want to add a dependent, and now workday will do all the work involved with that. And workday can charge a premium for that. So why is it input coffee and output code? The whole idea is that you now have software engineers that can build products on top of these digital or digitized cloud-based filing cabinets that now do the job that the end user of that software product did in 1960, 1970, 1990, 2000, 2010, 2023, 2024, but 2024, 2025 onward, now you have software agents that are effectively doing what for 65 years had been human work.
7:38 Speaker 3 Yeah, and that sounds really important. So I want to underscore that. You kind of went through the eras you have the non-software era, then you have software, then you moved into cloud, then you have this financial services enabled cloud era, and now we're in this new era. Can you talk about how, as we chart through these different eras, the scale, maybe in this new era is fundamentally different?
7:58 Speaker 0 Yeah, it's true, completely different because it's really comparing wages to software. And to kind of take an example from a completely different field that also has no software market. Um, there are about 4.7 million registered nurses in the US. Average wage for a nurse is a little over 120,000 a year. Um, so it's it's a high paying profession. And that means that the annual nurse, not software market, but wage market is over 600 billion dollars a year.
8:27 Speaker 3 Wow,
8:28 Speaker 0 which is a lot. And the worldwide software budget, uh software market is under 600 billion dollars, not just America. By the way, that nurse figure is just for the US. Of course, there are nurses in the UK and France and Angola, like every country on earth, right? The labor market is enormous, but what is the dedicated software market for nurses, like probably zero because nobody took the time to develop software. And you could try developing software, but it's just there's no budget.
8:55 Speaker 3 Economics didn't make sense.
8:56 Speaker 0 Exactly. If every hospital in the US currently has a nursing shortage and they want to hire a nurse, or maybe you're in Minneapolis where there's a giant Somalian uh expat community and you need nurses that speak that language, how do you find them? And like it takes three to four years to actually get trained as a nurse. Now you have a software product that can deliver not everything that a nurse does. Of course, it can't be a phlebotomist, you can't uh perform CPR, but it can call you the night before your colonoscopy and say, don't eat food. And it can say that in 45 languages, and it can actually have a conversation with you. So that's a labor example, or I want somebody like going back to financial services land, um, NetSuite is used by I think it's uh 70% of companies that go public are on Net Suite.
9:42 Speaker 0 It's some very very low. Sure,
9:44 Speaker 3 I've heard a podcast ad with that exactly. Yeah, I mean it's maybe it's higher.
9:48 Speaker 0 Something on that order of magnitude. It doesn't, if I'm wrong by by 10 points, it doesn't uh belittle the point. But uh the key thing is you always have people that are paying you late. Like you'll look at your accounts receivable and you'll see, wow. Like a bunch of customers owe me millions of dollars. And that's what you'll see when you look at your financial statements. And then again, this is where humans take the operation on that information. I have teams that are in collections that will go call you and remind you to pay or I'm going to cut off the product. That's an operation that can now be done within Net Suite. They haven't done this yet, but I'm sure they will. And versus charging for like the filing cabinet, now they could say, well, we know that you wanted to hire five collections people. We know that you pay those collections people $80,000 a year with benefits and everything else.
10:34 Speaker 0 We know that it takes a year to train them. Now our software product can do that not for $80,000 a year, but for $2,000 a year. And that's incredible. So it really is augmenting. I mean, the question will is actually how will the customer think about this? Will they really say, because once they start paying Net Suite 10 times more than they paid Net Suite last year, they're like, wow, my software budget has ballooned. I got to cut my software spend. Or will they say, wow, I'm saving so much money because as opposed to these five job openings that have, you know, I'm waiting to pay 400,000 a year for these five collections people. Now I can pay $10,000 a year to Net Suite and I'm paying more for software, but less for labor. And that that part, it's very, very new. But a lot of the hypergrowth that we're seeing in this category of company is because they are really moving into the labor market and less the software market.
11:21 Speaker 0 And that nurse example is kind of a prime example of that.
11:24 Speaker 3 Definitely. It'll be very interesting to see the appetite. If you think back to the app store early days when people were so uh reticent to pay 99 cents for an app for arbitrary reasons, but just because it was what they were used to paying. So I want to talk about uh pricing and whether companies are adapting and also consumers, but super quickly, as we round out these eras of clouds. Maybe it'd be interesting to just have you touch on how the previous eras have actually set us up for this one, right? You mentioned things like PeopleSoft or Quicken or Zendesk and all these companies that are capturing data. How are we uniquely set up now because of the previous eras?
12:02 Speaker 0 Yeah. So I would argue that if we just went straight to AI in 1960, like this just wouldn't have worked because you still need human input to go like collect the customer information, put it there. Like we have everything built out where right now this isn't an API. But if I want to go answer a customer question, the question's already on the computer, it's already on the internet, it's already in a database, like all of these things have set this up to be the ultimate platform. So this is why we love investing in what I would call systems of record. And a system of record is just something that has every single piece of minutia that runs a business. And it could be the strangest business that you can imagine. Like if I'm running a laundromat, there is actually dedicated laundromat management software. All of these systems of record that have popped up for all sorts of different businesses and all sorts of consumer use cases.
12:53 Speaker 0 Like the fact that now the mainstream form of communication for many, many adults and children is texting and email. Like it's not voice, it's not calling your parents that you haven't talked to in a long time. It's all in the cloud. Now you can perform these operations. So, you know, I would almost say there's been a 60-year period of digital digitization of physical things, putting them in the cloud so you actually, you know, what why is the cloud part important? Well, again, if this were the mainframe era of the 1970s, how would you get the AI to go like that's in some Google server somewhere or like thousands of servers? How would it actually have access to all of the information that's in some server basement in Indiana? Like that, that's really hard to do. So all the data is here in the cloud, all the compute is in the cloud, and now you just kind of mix them together.
13:42 Speaker 0 So the fact that systems of record for so many different types of businesses and so many different kinds of consumer use cases are now widespread. And you know, hundreds of billions of dollars of company market cap has been created from these systems of record, either horizontal or vertical, and a vertical one would be something like a toast that vertically runs a restaurant and a horizontal one would be like a Zen desk that just does customer support software in the cloud, but it does it for every different type of company.
14:11 Speaker 2 Building on that, if you take the toast example, like start with the cloud wave, move to the financial services wave. It was initially a hypothesis that these vertical SaaS companies would make a lot more from financial services. Fast forward to today, 80% of toast revenue is payments, insurance, like all sorts of financial services versus software. And so as we think the next wave is going to be, you know, if you're operating toast, or one of my favorite examples is Mind Body, which runs, you know, fitness studio software. It does scheduling for employees, it's a CRM. They also make a lot of money from financial services, but they still need a lot of people beyond, you know, you're never going to replace the yoga instructor, but you've got your financial back office. You've got people answering the phone to answer very basic questions. Like all of that can start to be done with AI.
14:56 Speaker 2 And I think the most bullish version of that is all right, you then don't need to hire people to do the tasks that are not human-facing that AI can do better. You know, is that going to increase software revenue 2x? It could potentially increase it 10x, depending on how the customer views it and how much they're willing to let their software budget bleed into their labor budget.
15:16 Speaker 1 I think part of the challenge or or the potential for disruption is that the pricing model may need to change pretty significantly, right? You talk a bit about this in your piece, Alex, where you know, Zendesk today is charging on a per seat basis. But if you're if you're actually eating away at some of the labor, you can charge per the output of work, right? And how does that sort of create the potential both for increased ACV, but again, creates uh, you know, potential disruption for a lot of these larger incumbent players given their existing kind of pricing structure.
15:45 Speaker 3 Can we actually talk about that example from Zendesk? Like, what is the difference between the software component and the human component?
15:51 Speaker 0 Well, so it's it shows how stark the difference is. So most companies, like Salesforce charges per seat. Zendesk, at the time that I wrote my piece, they might have changed their pricing a little bit, but it was $115 per seat per month. So imagine that you have a thousand people that work in a call center or really more of an email-based support center, and they use Zendesk and Zendesk then profits as you hire more people because a thousand is better than a hundred. So $115,000 a month is about $1.4 million a year in spend on software for Zendesk. And Zendesk, I think has about $2 billion in revenue, something on that order of magnitude of annual recurring revenue from all these seats that are paying every single month. And of course, they want their customers to grow seats. Now, if you assume that each seat, how much is that person paid?
16:39 Speaker 0 How much does their healthcare cost? How much does their, you know, stipend for commuting and you know, yoga benefit and all these other things that the company throws in, um, you know, maybe it's $50,000 a year per person. So what's $50,000 times a thousand? Again, a thousand seats, that's $50 million. So you're spending $50 million on people. Um, and by the way, it's very hard to hire and train these people. So I think one misnomer is like, oh, AI is terrible, it's gonna end all employment and it's gonna be chaos. I always like to point out that when the United States was founded, um, 97% of people in the US were farmers. And most of them were put out of jobs by things like the tractor, but that's actually fine because they moved on to other kind of productive parts of labor. And by the way, the the average life expectancy was a little over 30 back then.
17:27 Speaker 0 Right. So I feel like what we have things have gotten better. Like I like uh I like penicillin and all the other benefits that we've gotten and fertilizer and all these things that have made our lives better. But so you have $50 million a year for people. Um, and then you have $1.4 million a year for software, which one's bigger? And this is the concern is that on the intermediate basis, and Zendesk is actually very lucky because they got they were a public company and they got taken private. So two big uh private equity firms bought it. And they're actually working on this right now because they're like, uh oh, if we make AI really good, then the customer that has a thousand seats might cut down to 10 seats because there are two forms of AI tools. I mean, there's more than two, but the the common example that we talk about is there's autopilot and then there's co-pilot.
18:17 Speaker 0 So copilot is like a productivity enhancer. So I'm trying to figure out how do I answer Angela's query. It has all these questions. I just got hired yesterday, I don't even know where the bathroom is. What do I do? And then it kind of gives me it's like, hey, we think this is the right answer. It makes Angela so much more productive in her job, and that's great. Um, autopilot is like Angela quit yesterday, and we need somebody else to go answer emails because it's you know, Black Friday. What do we do? What do we do? Okay, now we just kind of throw the tool at the customer directly and have them answer the questions. And that's the big danger. It's like copilot is actually a danger for revenue as well, because why do I have a thousand support reps? I have a thousand customer support reps because they can only answer 10 questions a day
19:04 Speaker 3 and I get
19:05 Speaker 0 10,000 queries a day. So it's just basic math. If now with Copilot, each one of my reps can answer 100 questions a day, I only need 100 reps. So now Zendesk lost 90% of its revenue. And then by the way, if autopilot becomes a thing and actually works very, very well, then I need nobody, and then therefore I sell no seats if I'm Zendesk. So it's a real, real, it's it's kind of both defense and offense for these companies to figure out what the hell to do. Because if they play offense, they could maybe 10x the revenue because of the example, like it's like 50 million dollars for people or 1.4 million for I'd rather get 50 million from Zendesk than 1.4 million, but they're probably not gonna be able to take all 50 million, right? Because ultimately, like the cost of delivering these services is very low. The moat is not that high. It's higher for people for companies that have the system of record, I would argue, because all of your past correspondence with all of your customers is in uh Zen Desk.
19:57 Speaker 0 VOM Salesforce, like every single communication that I've ever had with any of my customers, my pipeline, everything, it's in Salesforce. It's hard to yank that stuff out because everybody's using it. And it's not like this binary thing where like tomorrow we're all autopilot. Like we're going to see a lot of these co-pilot tools. They will run on the systems of record. But even as they're running on the systems of record, now I need less seats. So that's why, you know, Salesforce, it's a 200 billion dollar plus public company. If they don't do this right, they could lose all of their revenue or most of it. If they do it really well, they could 10x the revenue. And it's like, where is it going to go?
20:34 Speaker 1 As early stage VCs were very excited about this. Yes, yes. It's all upstairs. Is
20:38 Speaker 2 the question, right? More about finding the next entrepreneurs versus the uh the fate of the internet. Well, it's an
20:43 Speaker 1 interesting moment in time for kind of, you know, enterprising young founders to kind of reinvent the model and charge radically different, you know, because many of these incumbents aren't going to evolve. Yeah. They're not going to change their pursuit pricing and they risk disruption.
20:56 Speaker 3 And is that the wedge, right? If if you're a startup and you're trying to figure out how do I enter the market, especially when you have companies like Salesforce, which have the system of record, they have all the data, they have all the customers. Is that the wedge where you say I'm going to undercut and I'm going to charge a tenth of the price, even though I still have great margins because I'm entering the labor part of the equation?
21:15 Speaker 1 I think so. Yeah. I mean, I wrote a piece recently that I called the messy inbox problem, which is sort of my way of describing a wedge strategy that we're seeing across lots of different industries. And the idea is basically that there's a class of founders that are starting building software products to solve a lot of the, you know, what was historically judgment intensive work. You know, in lots of different industries, there is some sort of human administrator's job. It is to basically extract information from a wave of unstructured information, whether it's emails, faxes, you know, transcribing phone calls, and then put that information into one of these downstream, you know, systems for record. It could be an EMR, could be an ERP, uh, it could be a CRM system. And historically, you know, that work lived kind of upstream of any of that software, right?
22:02 Speaker 1 Because it was the human's job. And software couldn't do that as Alex described. Now we're seeing, you know, companies sort of wedge in and replace again that messy inbox problem with software and slowly begin to eat away at all the kind of downstream workflows. And over time, I think that the thesis is that while that initial wedge uh is highly differentiated, you know, against the human, uh, it it really is the opportunity to kind of eat away at everything else and become the kind of new AI native, you know, system of record. Um we have a company, you know, it's an example called tenor that is doing this in a healthcare context. So, you know, the problem that they're solving specifically is uh is around patient referrals. So you go to your general practitioner, they're referring you to a specialist, could be a dermatologist or an imaging center today.
22:49 Speaker 1 They're often faxing your medical records, and it is somebody's job to go physically to the fax based and you know, take that physical fax, re-enter that information into that to the EMR system. Tenor has trained a model against, I think, four million, you know, healthcare specific documents and now can basically extract all that information about the patient programmatically and have effectively begin to solve, you know, this sort of patient intake problem. And they're able to reduce now about 90% of the admin costs, you know, of that patient intake before the patient is actually seeing the the clinician. Um they again wedged in with the messy inbox problem. And over time, they're now eating away at things like scheduling and uh eligibility and benefits. Um, and you know, over time we'll see if they become the kind of core AI native system record.
23:36 Speaker 3 And something you're pointing to there is also the defensibility of it all, right? So you can get the wedge maybe through pricing, but how do you actually protect your customers? And I think one place people jump to is okay, you need your own models or you need some sort of proprietary data. Is that the defensibility of you know, this future, or how do you think about that?
23:54 Speaker 1 Sort of this distinction between differentiation and defensibility. I think AI is an incredible catalyst for differentiation, right? Solving the messy inbox problem with software is a thousand times better than the human doing. It's not even on the same kind of playing field, super differentiated way to sort of wedge in and again kind of own the downstream workflow. Is that wedge product alone defensible? I would argue no, right? Today, it feels like magic to the providers that they're working with. But I think that capability is gonna become commoditized over time. They may have an advantage because they've trade a model, you know, for now, but I think that is ephemeral. I think the defensibility comes from again, owning all of the downstream workflows, deeply integrating themselves into every other system that they have, effectively owning that sort of you know, core kind of end-to-end workflow.
24:40 Speaker 1 And I guess the hot take would be, you know, that moats still matter. And a lot of the motes in software today are are the same that they've always been, right? So becoming a system of record, you know, having a network effect, becoming a platform, having virality baked into your product, deeply embedding yourself, you know, in you know, into the existing system. So it's hard to rip out. Like these were all of the heuristics that we would always have looked for in software. I think that, and they're still true today.
25:06 Speaker 0 So I agree with all of that. The other way to think about this is why did software start with airlines? Well, people, everybody traveled, not everybody, but a lot of people traveled. Airplane tickets were very expensive. And this was kind of a pittance for them, and it made so much sense versus having like throngs of filing cabinets and gophers, and it makes sense to go pay, you know, hundreds of thousands of dollars in 1960s money to go buy some giant IBM mainframe. And this is kind of why I brought up that financial services example. Like software for restaurants did not make sense. It just wasn't a problem to be solved, and the market wasn't big enough. And you made the market big enough, once to Angelo's point, you threw in like payment processing and insurance, all these other things that they were paying for anyway. So you can either try to come up with a wedge, as as David mentioned, and then figure out how you kind of expand your wedge and make it defensible and become the system of record.
25:59 Speaker 0 The other thing that you do is you just find like things like restaurants in 1980 that just like have no software and needed no software, wouldn't pay for software, but their labor budgets are enormous. And uh sometimes you don't know what these things are, but like the example that I would give here, what is the incumbent software product for compliance officers at banks and financial institutions? Like I don't like Excel, Word, uh Microsoft edge browser looking for bad things. Um what does a compliance officer do? I mean, it's it's now in the news a lot because of this de-banking thing. And I found this on the Bureau of Labor Statistics, the fourth fastest growing job in America is compliance officer. There isn't an incumbent software product that they use. Every bank and financial services company on earth, um, they're all hiring for compliance officers.
26:44 Speaker 0 It takes a long time to train them. And what if account openings goes down? I don't need as many. What if account openings go up? I need more. Um, it takes like in some cases, like a month to open a business bank account at a bank because the compliance officer or the compliance officers are backlogged. What if you deliver that via software? Right. And there is no software. There is no incumbent that can now like add this AI module in the same way that NetSuite was an incumbent for accountants and financial officers at companies where they can add a module for AI work of collecting money. Um, so you find these other areas where there really isn't an incumbent or the incumbent is like it's like Microsoft Excel. But these sometimes these are just so bizarre. Like I wouldn't think about like compliance officer until I saw this like random report from the Bureau of Labor Statistics that showed Maticurist is number one.
27:32 Speaker 0 And then, and that's a hard one to have AI do. Um, and then number four is compliance officer. And then you kind of look and you talk to banks, they're like, what software do you use? And it's like Excel. Again, giant labor budget, not enough people that are using software. That's one of just like you can not have to worry about the incumbent layering in something. And it still is a wedge, but you could probably turn it into a system of record. And why, why were no venture-backed companies or even non-venture backed companies built in this space? It's just like, well, I can't charge that much money. It's just not a big market in the same way that like there was no restaurant software market in 1980. It's the exact same reason, but there's enough budget there to go fill this like very, very pressing need.
28:15 Speaker 2 I think one of the one of the most interesting parts of our job is every time there's a new technology shift, we have to challenge every investment thesis that we thought was not gonna work. And so many of them now are going to work. If we come back to financial services, there's a lot of pretty terrible systems of record where smart people have tried to get them ripped and replaced and just it was just not gonna happen. And my new conclusion with AI is not that they would never do it. It's that the replacements were 2x better. They weren't 10x better. And so if we come back to compliance, lots in Excel. But if like, for instance, you've probably read the $4 billion fine by TD and transaction monitoring, right? And so they said on an old transaction monitoring system mactimize as one of them. They should probably get one that throws up fewer false alerts, but they were trying to clear a backlog of several tens of thousands alerts.
29:03 Speaker 2 They can't hire enough compliance people. So now an interesting wedge in is we'll provide you with all of these agents. And oh, by the way, we also have a much better transaction monitoring system that is actually gonna fix the problem. So this labor plus software bundle also helps the sales process and then helps the defensibility because you're really solving the major problem, which is better software and also I can't hire the people.
29:25 Speaker 3 Yeah. And I think that's such an important point because you all have pointed out different areas where quite frankly, the labor is not there. And so as we're talking about software disrupting labor, the natural question is, okay, so what happens to all these jobs? But maybe we can talk about both that and the flip side of what new jobs are created, because you know, even if we talk about the previous arcs, we saw what product managers, UX designers, social media managers, those were all remnants of the previous era. Um, how do we think this will kind of shape up in terms of new jobs and then also how existing jobs change?
30:00 Speaker 0 It's always hard to prophecy these things because in 1789 or something, it would be hard to say like what will these farmers be doing? Uh post-tractor sitting in a room talking. Exactly. Talking with these these electronic microphones and this like amazing fire.
30:15 Speaker 3 The fire in the
30:16 Speaker 0 sky here that we have in the basement. Exactly. So you know, what what was a nurse, right? It's like medicine was bloodletting leeches and prayer. So it's obviously changed a lot. So on the one thing that I think AI cannot do, um, and in fact, if anything, it kind of commoditizes, like if you have an AI sales rep, like Salesforce, why do I need seats for salespeople if AI is doing selling, but AI cannot build a relationship with somebody over golf? So I I think the in-person, like the things that only humans can do, that uh that skill set might go up in value tremendously. It obviously makes sense that I mean, like at the far extreme, I I talked to somebody who believes that in the future, the distant future, there will only be two jobs. You either tell a computer what to do, or you were told by a computer what to do. Um, and there's a whole set of things where people could be a lot more productive in whatever job they're doing, or kind of like part-time work when you have this little coach by your side saying, do this, do that.
31:11 Speaker 0 But I think the the human connection thing is almost the more most important because if I think about every other era of like a new communications tool, um, so imagine having the first telephone, you know, Alexander Graham Bell invents the telephone. Nobody has a telephone. Like, how do you scale this network? You know, you get one, and the only person that calls you is your your mother saying, why don't you call me more often or something? So then the first telemarketer shows up and then starts taking advantage of the fact that you have a phone and they can rather than like go horse and buggy to your house, they can try selling you something over this like old fashioned telephone. Um, that was a very advantageous place for the that first telephone telemarketer to be. Or the the history of Sears Robo, Sears is really fascinating because even though they they had the Sears Tower, they had these giant stores, it really was the first like giant mail order catalog.
32:02 Speaker 0 So they were the ones that kind of figured out how to use the US postal service. Faxes came out, people started sending unsolicited faxes. So the reason why I bring all this up is you can imagine a world where like all of like AI is selling you everything, pushing you everything, and then right now it's a novelty and it works really well. But once it becomes so mainstream that everybody's doing it, it's like this Yogi Barrier expression, like it's so crowded, nobody goes here anymore. You can imagine like the need for actual human connectivity to go up dramatically. It's followed this pattern of like once it gets so crowded, like the somebody that's doing something different, in this case the old fashioned way, like might be more valuable.
32:42 Speaker 2 One of the ways we think about it is like all of us have some percentage of our job that's wrote tasks that could be automated with AI. And I think we strongly believe that at least every white collar job is going to have a copilot. Some might be fully identical, you know, back to our L1 compliance reviewers. And so if you imagine all of us not doing any menial tasks or anyone and focusing on the human connection, the most creative parts and having all of our days to spend on that, like what might be enabled. And that's a pretty exciting way to think about it.
33:12 Speaker 3 And as we think about the companies that can be created in this wave, I'm so curious, because it does feel fundamentally different as you three are assessing companies. Are there a new layer of metrics that you pay attention to again using a previous wave? Maybe we got social media and all of a sudden we're thinking of daily active users. That was kind of a key metric that people started to pay attention to in this new wave. Like, are there new metrics? Are they the same metrics that matter? Is it too early to tell? I'm so curious.
33:39 Speaker 0 I think it's actually the exact same metrics. It's not like, oh, it's AI, so therefore future profits don't matter. It's like it's the present value of future profits. And that really comes down to like how many customers do you have? Do you retain those customers? And then how much gross profit do you make per customer, and then how much overhead do you have? And I don't think any of that changes. So I mean, the reason why social networks were interesting is we knew that the customers retained, right? But it's like, will people pay for it? Will it make money? There was this open question. And therefore, there was this, if you will, alpha of, oh wow, like it uh we we call this the smile curve, which is very rare, where like more people, obviously 100% of people use the product on day zero because day zero is when they installed it. But then people stop using it on day one, day two, day three.
34:25 Speaker 0 And normally most products, they just have exponential decay. So by day 200, you know, of the hundred people that downloaded on day zero, zero people use it at the end. Um what's interesting is things like Uber or Facebook, where again, 100% of people use it on day zero, then it drops off on day one, day two, and then it picks back up. And then it plateaus at maybe 50, 70, 90% of the original starting bunch. And that's so rare. But then the question was will Facebook ever make money? Like, oh, they won't make money because it's free. But they figured out that advertising was very valuable. Um I I think the vast majority of things that we're seeing right now just monetized via subscription. So it's actually very clear how they make money. And the DAU thing is actually just as useful today as it was before, but it's like the money part is almost automatic.
35:10 Speaker 3 Like
35:10 Speaker 0 the thing that was unique about the the internet era was it's like, oh, like get big and then monetize later. And we're not seeing as many of those, but I don't think any of the fundamental isms of evaluating a business have really changed. I mean, if the the only thing that's kind of more dangerous is since AI can now write software, it's just so much easier to spin these things up. Whereas to build something and scale it out, like you know, the reason why friendster failed, like friendster should have been the social networking winner, but their servers couldn't stay up. Um and MySpace should have been the winner, but they couldn't hire good engineers. Like there are all these different reasons that are not relevant today because the technology stack is so different. But again, it's you know, present value of future profits, that's unchanged.
35:57 Speaker 2 I think one thing, and this is not a business metric, but that's changed is just potential market size. And we talked about how it scales in the large vertical SaaS markets. But if we look, so there's something in the in the US called the uh North American industry classification system, the NAICS codes, and there's 600 of them and they'll classify industries, how many companies are in there, what's their what's their labor budget? And there's a whole host of industries. Whereas if you looked at them before, you say, well, there's a thousand potential buyers, maybe they'll pay a thousand dollars a month for my software service. That's 120 million dollar market. That's really not that interesting if I'm gonna build a venture-backed business. Now, if you think you can layer an AI, replace some of the labor budgets, those markets get dramatically bigger. And so I think the different pockets of where software can be built, these niche markets that weren't that interesting, are now potentially very interesting.
36:47 Speaker 1 I think the other dimension that we're seeing kind of pitches for is, you know, are you selling software into the incumbent industry or are you building kind of the full stack version? Right. And we Alex wrote a bit about this in in kind of our brick inside the gate, you know, which is sort of like the evolution of private equity kind of in an AI context. And so you think about an area like professional services or in legal, you know, for example, you know, the challenge that a lot of these law firms have is that they're charging on a per hour basis, right? So if AI can do what used to take three hours and three seconds, you know, where does the revenue go? And so we're seeing some people pitch the full stack kind of AI native law firm, which might have a totally different cost structural, you know, to Karath or you know, one of these big firms. Um or there are other areas within professional services that are much more aligned, you know, to benefit from that efficiency.
37:35 Speaker 1 So for example, we have a company that we we haven't announced yet. So I won't, I wouldn't mention the name, but that is solving a lot of the workflow challenges in uh in plaintiff law. So they operate in both employment and personal injury, where in that model, unlike on a per hour basis, they're charging on a contingency model, meaning they don't get paid unless there's an outcome or a settlement of the case. And in you know, personal injury, as an example, for every hundred leads that these lawyers get, they take one case. You know, and there's a ton of again, that messy inbox problem of sifting through, you know, medical records or employment documents and having to essentially value, you know, or quantify the value of each case that that they'll take on, because any any case that they that they take is essentially an investment of their labor. Um and so what this company is doing is essentially programmatically helping solve that sort of intake challenge, that messy inbox problem to help automatically qualify the value of those cases.
38:27 Speaker 1 And then works essentially as a co-pilot, you know, for the lawyer to draft a medical chronology to create a demand letter to file, you know, a complaint and basically walks through the entire sort of pre-litigation and and litigation process that essentially allows that lawyer to take on, you know, 3x or 4x the number of cases. But again, that that the value that that software is delivering to the practice isn't so much, I mean, it is also reducing labor costs. So one way to do it is, you know, you just have fewer, you know, lawyers and the and the same amount of revenue, or in this case, I think what's going to happen is it's going to significantly grow these practices. And in this case, they're actually passing the cost of that software to the end client in the form of a technology expense, which they often had done historically. And so the value that the software is delivering to each of these firms is super aligned, you know, to the impact that it's having on the business.
39:15 Speaker 1 And so I think, you know, the more clients you can take on, the more people that can pay for the for the software. And so on a per firm basis, there's a significant kind of revenue expansion opportunity. I think that's sort of an interesting tension that you'll see across industries where, you know, does the eye help, you know, by reducing cost? Is it better better to build a full stack version or sell the software in? And I think there'll be successes in both dimensions, but it's something we're seeing more of.
39:41 Speaker 3 Yeah. And one follow-up there, as you're talking about the cost being passed along to the end user or buyer, is that just net deflationary as this permeates across the system? I know it would take time, but uh eventually if you see more competition and more people creating these AI based uh labor products and then people. People are competing on price and all of a sudden, you know, to take on a new case. It's no longer $5,000, it's 500. You know how you said that in the previous era, the open question was can we make money? Can we monetize? Is that an open question that you know this over time just becomes deflationary and firms can't charge as much?
40:19 Speaker 0 I think technology, if it's done right, is always deflationary because you get productivity gains. So I think for sure, I think the defensibility point is the is the most relevant one that we struggle with a lot, which is wow, it's so easy to build one of these things because the number one use case, it's almost like a recursive thing. It's like, well, what's the best you who which profession is using AI tools the most? It's probably the tech people that actually build tools. And that's where things like cursor have gotten so popular. But what if there are 50 companies that end up doing the exact same thing? Um that's that's the hard part. But I think I I can't see a scenario where prices are more expensive than humans or where prices don't just keep going down.
41:04 Speaker 3 Um significantly.
41:05 Speaker 0 Significantly. And that's the history of technology in a nutshell. I mean, again, it's like the hundred megabyte hard drive in 1960 that weighed like tons, literally tons, uh, was probably a million dollars or something crazy. And now it's just comical, like these little, I bought some for Cyber Monday. I bought all these like little micro SSDs for like a terabyte was like $10. It's just incredible. So I think that is an inexorable process for technology cost in general. And then you also get new use cases where I mean, I like this right of the supply demand curve. This is a really interesting use case where it's like there just wasn't demand where there was supply. There's a lot of supply to do something for $2,000 an hour if I want to, you know, file a trademark with the leading trademark attorney.
41:50 Speaker 0 So the trademark market or the patent market, like that might be very small because it costs too much. But now if it only costs $5, wow, maybe everybody does
41:59 Speaker 2 it.
42:00 Speaker 0 Um those or like translation is one that I find fascinating, where it's like it just doesn't make sense if you're a small company to go translate your introductory video into like 45,000 different languages that have ever existed. Like why would I translate it into ancient Greek? But you know what? Why not? It's free. So you have all these other things that just kind of expand the market because the cost has dropped so precipitously.
42:26 Speaker 3 Where do you guys want to see more builders applying themselves? You're obviously seeing a lot of companies, a lot of people excited. Also the incumbents are clearly excited about getting in on this wave. Is there an area that you'd like to see more attention being put toward?
42:39 Speaker 0 Obscure is good. I mean, we we love it when somebody walks in, has had um a decade or more of like obscurity. Uh they they they serve some weird job or they were at like something that nobody's ever heard of, like in like the farming industry, the mining industry, the whatever industry. And then they actually have an insight that somebody else doesn't, and they actually know the potential of AI. Because the other thing is that it's very important to know that the technology is not ready for autopilot for a lot of these things. It's just like the use cases are too complicated. The integrating the different pipes is too complicated. So kind of overshooting early, like they're gonna be a lot of failures because there inevitably are in every technology revolution, not because the idea is bad, um, but it's just like it's not good enough to be a hundred times better. So I think it's like obscure, and then finding the ones where like at least for right now, it's not like it actually the technology is good enough for the obscure use case at hand.
43:33 Speaker 2 Yeah, I'd say also across there's there's many industries of this, but financial services and insurance have a host of old system and old, like 30 plus year old systems of record that now can be made 10x better incorporating labor, redoing them in a workflow. And so deep knowledge of those areas, like we have transaction monitoring at Sardine, we've got a mortgage loan origination system in VESA, we've got servicing. There's a lot, we've got a couple in insurance. And so entrepreneurs that really understand those space and can bring AI thinking there. I think it's a big opportunity.
44:04 Speaker 1 Yeah, look, I think we'll continue to see lots of entrepreneurs, you know, wedge in with the messy inbox problem across lots of you know niche vertical industries. I think it's also we're still on the lookout for you know horizontal software, you know, kind of AI native versions, you know, selling into the sales teams, marketing, you know, product management, analytics, CFOs. Um, you know, in in that in those categories, you often do have a large incumbent software competitor. And so that's sort of the tension. You have to understand the market structure and how likely it is for that incumbent to change their pricing model and build kind of more AI native features. But I think there will be generationally defining companies, you know, built in an AI native way in in horizontal software as well.
Previous Building Substack: Reinventing Culture through Subscriptions Next Marc Andreessen: Can Tech Finally Fix Healthcare? Page updated August 25, 2026 · transcript and summary are machine-generated
