The best argument I’ve heard for why AI won't take your job with Box CEO Aaron Levie

This podcast is brought to you by Atlassian Rovo, the AI that takes your team from AI novice to AI native. Hey, welcome to Platformer. I'm Casey Newton. Recently at Platformer, we announced a new push into original journalism. We've always done interviews with tech leaders, but we wanted to experiment with extending those conversations into new formats like audio and video. So for the next several weeks, we'll be bringing you a series of conversations trying to make sense of what's happening on the ground in Silicon Valley today. I'm bringing in people I've known for a long time and people I've only met more recently to try to get a handle on a very confusing subject. What is the future of work in a world where the AI systems keep improving? I have to tell you, it's a really strange moment to be covering this industry. The companies building AI are convinced that they're on the verge of automating away most knowledge work. Meanwhile, the companies buying that AI are way less sure about that. And then on the worker side, you have a split between people who say that their jobs have already transformed quite a lot, and other people who say nothing has really changed. And also, they kind of hate the way that their bosses are talking about it. Meanwhile, layoffs keep arriving, with AI being cited as a partial cause, even as the executives ordering them insist that AI is somehow going to create more jobs than it destroys. The stock market thinks one thing on a Monday and something completely different on a Friday. Nobody seems to agree on what's actually happening, which is usually a sign that something is. So we're going to do what reporters do, which is talk to people. Over the next several weeks, I'm sitting down with founders, CEOs, researchers, and operators, the people whose day jobs require them to have a working theory about where all of this goes, and ask them what's really happening. We're also going to do another thing good reporters do, which is bring you data. Each week, we're going to kick off the episode with fresh numbers from the most interesting research, surveys, benchmarks, and other sources that we can find to help tell the story of how work is changing. And I did not want to do that first part all by myself. So to kick us off, I'd like to bring in platformer fellow and Gen Z AI correspondent, Ella Marchianos. Ella, how are you? I'm doing well. I've had a beautiful morning reading news stories, browsing Twitter, journalizing. I don't know what else you'd expect me to have been doing this morning. I mean, that's exactly what I expect you to do each and every morning, because that helps you, you know, complete the task that I have given you, which is each week bringing us a story that helps to advance our understanding of AI and work. So I am very excited to learn what are you bringing us this week? Yeah, so this week we've seen two pretty interesting large surveys about how people are using AI at work. I don't know, it's like an amount of data on this that we haven't necessarily had before. And one thing that really struck me in both of these surveys, one's from the Financial Times, one's from Gallup, is people who are higher earning in their jobs, people who are in leadership positions at their jobs, are actually using AI a lot more of the time than like more junior workers or people who are earning less money. So the big headline data from the Financial Times, for example, is they split up workers into like deciles of income. The like top 10% of income, 60% of people are using AI most days. The bottom 10% of income, 14% of people are using AI most days. And this is across like all sort of knowledge work jobs or it's every kind of job? Every kind of job. A lot of the sample is knowledge work as far as I can tell. Got it. Okay. So what do we make of this? Does it seem like it's as simple as if you use AI more, someone gives you a raise? Or is it that if you are a higher earner, maybe you're in this sort of job where you just sort of understand, hey, I really need to be using this stuff. Yeah. So a few things might be going on. There's the simplest deflationary hypothesis, which is people who earn more money just have more computer jobs. They're writing more. They're writing more code. And because this contains a pretty broad variety of professions, a lot of which are computer jobs, it might not just be that. Another thing that is brought up in the Financial Times article is just people who are in general in a higher income percentile have more tech literacy. They will know more about these systems, what they're useful for. They'll be less likely to be afraid of them. And so they'll use them more often. Okay, so let me just ask one question about this study before we move on to the other one, which is like, do you see anything actionable in here? Like, do you think that based on this, workers have an incentive to like, use AI more in the hopes that it will mean that they get higher pay? Or are we just sort of observing that like, well-paid people seem to be using a lot of AI? Right now, I think we're just observing that well-paid people use a lot of AI. Like the study authors have specifically said they want to do a follow-up that tracks like promotion patterns. And currently, like the age of these systems and the data is such that we can't show causation. Got it. Okay. Tell me about the Gallup survey. Yeah, basically they, among other things, are asking people whether AI makes them more productive at their jobs as well as how often they are using AI. And one big thing is when people were asked if AI makes them more productive at their job, 71% of people who said they were in a leadership position said that AI made them more productive. Whereas like only 54% of people who said they were individual contributors said it made them more productive. And, you know, it has this graph where, like, as you go down the totem pole, people are a little less optimistic about whether AI makes them more productive. Yeah, like, the closer you are to entry level, the likely you are to say AI is bullshit, right? But then, like, and the closer you get to the C-suite, like, the more likely you are to say AI is everything. I wrote a piece earlier this year, the AI productivity paradox. It's about exactly this dynamic, and the finding was exactly the same. Um, what, uh, hypotheses, uh, did the, the, the study authors have for why these higher earning leaders are using AI more or like getting more out of it? Yeah. Basically the top two hypotheses are that, uh, they will know how to use the tech and that they will use more computer, they will have more computery jobs, um, which is basically the same as the Financial Times. As the first study, yeah. Yeah. I have a third wild theory on this, which is if you're a more junior employee, you have less executive decision over how you're like less power over how your job gets done. And if you try something new and it ends up like messing things up, you are just like in way more trouble because you don't necessarily have jurisdiction. So you don't necessarily want to go to your manager and be like, how much should I be using AI for this? And especially then be the first AI adopter when something goes wrong. Right. That makes a lot of sense to me, right? It's like scary to be in that entry level role. And I think often people don't really feel empowered to say like, hey, like, would you mind if I bring my own tech stack into the enterprise? Like most people are just going to wait to be like, you know, given their their notion login or whatever. All right. Well, it's an interesting set of findings. Any sort of like final conclusions about like what this means or like if you're a worker, what you think you should do with this information? Yeah, I actually I want to bring up kind of a non sequitur, which is there's another finding in the Gallup survey that I thought was pretty interesting, which is how much people report productivity gains by sector. And some of it, so the greatest gains are in managerial and healthcare, which you would kind of expect. The lowest gains are in service, but then like the second lowest is office and admin support. And, you know, there's kind of two things here. One is there is a clear reason why you might see lower adoption, which is one, knowledge, and two, like some of your job in admin support is literally moving boxes around. You know, like chat GPT is not going to do that. Can't do that, yeah. But like I've worked as an office assistant before and, you know, like some of the time I was moving around boxes, but some of the time I was doing stuff like I open a spreadsheet and I check if people are listed twice on the HR spreadsheet. I check if people who are gone are listed on the HR spreadsheet. Like I port copy into WordPress. And, you know, that's really stuff where I'm not like Claude can't do this. Claude is not useful for this. In fact, like I can see plenty of roles within this sector where like AI maybe per unit time is more useful. And then another thing that adds a little nuance here is there has been a body of literature previously, like from 2023 to 2025, where people do trials within workplaces of how well, like how much AI boosts productivity, where typically the finding is that lower skilled workers, people with less job experience, have their productivity boosted more by AI. um and so i don't know these two anecdotes this anecdotal thing and then this previous literature thing uh it indicates to me that there's like some background evidence here that the difference is really in fact not just because of usefulness like there are places where more junior employees where people uh in like less knowledge work or heavy professions could be using ai more but empirically they aren't. They don't feel like AI makes them more productive. Right, they're choosing not to and something their managers may want to get curious about. Well, I think it's a really interesting set of stats to consider and exactly the sort of thing that I hope that we can bring to people each week during this series just to try to ground ourselves in even more data about what's happening in the world. And with that, I think it's time to bring in our guest and asked him about these and other matters. So after the break, we'll have Aaron Levy, CEO of Box. Not sure how to actually use AI at work? Most AI tools promise to save time, but they set outside your workflow. They're disconnected from how things actually get done, and they don't make your to-do list any shorter. Atlassian Rovo works where you work, across Jira, Confluence, and the rest of your stack, securely connected to your company's knowledge, context, and permissions. Turn meeting notes into Jira tickets. Draft campaign briefs and confluence. Instantly find the right docs without digging. This isn't generic AI. This is AI that understands how your team actually works. Less searching, less busy work, more progress. That's what an AI-native team looks like. Learn more about Rovo at rovo.com. Welcome back to the show. My guest today is Aaron Levy, the CEO of Box. Aaron was one of the first CEOs I met when I arrived in Silicon Valley in 2010. Box had begun in 2005 as a college project for Aaron when he was at USC. When I met him, they were a couple years into a pivot away from being a consumer business to the enterprise. and at the time the big story about box was its rivalry with dropbox which had stayed focused on the consumer and at the time dropbox was so big and buzzy that lots of people thought dropbox was just gonna win and box would disappear fast forward to today and dropbox and box stock price trades at about the same price but they have a new threat to worry about and some folks are calling it the saspocalypse the saspocalypse thesis is the argument that traditional software as a service the model where companies pay per person subscriptions called seats for software is going to go away or rather that it's going to be devoured by ai the logic goes something like this most sas products are essentially structured interfaces on top of a database and their value is in organizing the work that humans do. But if we arrive at a time where AI agents can do that underlying work directly, drafting a contract or updating a record or reconciling an invoice, then maybe you don't need the interface anymore. Maybe you don't need Box anymore. At that point, the buyer no longer needs 100 seats of a company like Salesforce or Box so that companies can log their tickets. What they really need is a model that talks to the database and a much smaller number of human supervisors. So that's the bear case for enterprise SaaS. Aaron, though, has a completely different point of view. He's been a vocal proponent for the idea that businesses like his are going to survive and even thrive in the AI era because he says agents aren't going to be enough. He thinks agents are amazing. Box has an agent. It's called Box Agent. and they're very excited about it, but he believes that you're actually going to want to have even more humans in your enterprise because there's always a last mile of work that the agents are going to be unable to do. I have to say, it felt like a very provocative set ideas to me as someone who is currently steeped in a world of AI CEOs telling me that automation is at our doorstep and is about to take everyone's job. So I'm excited to share this conversation with you. If you are somebody who has been really, really worried about automation, I think it'll give you a fresh set of ideas about why, if that happens, it will happen much more slowly, but also reasons why human beings might very well still have jobs, even good jobs, well into the future. So with that, here's my conversation with Aaron Levy. Aaron Levy, welcome to Platformer. Hey, good to be here. So, Aaron, you and I first sat down in 2010, and you explained to me. We were so young. We were so young. We were so young. You know, back then, you were, like, sort of early gray, but now you're just, like, normal gray. Yeah. And I think running a public company will probably do that to you. The problem is, actually, I've been like this for 13 of those years. So it would be one thing if this only occurred, like, the past six months, but it's actually when I was 24. It started sooner. Maybe there's more reasons to be gray today. Or maybe not. We'll get into it. But, you know, that first time that I met you, I have this kind of core memory because I truly had been in Silicon Valley for what feels like weeks when I came down to the box office. Did you come from, like, the Arizona Star Tribune or something like that? The Arizona Republic, that's right. I've been covering local government. And then one day I said, what's going on with computers? That seems interesting. Yeah. And now here I am. But I needed people to explain it to me. And that's kind of where you came in. And as I recall, you just kind of explained the software as a service business model to me on a whiteboard. It was super useful to me. Little did I know that wouldn't matter. So, yes. This is kind of my first question, which was like, if you were explaining your business to a reporter like that today, how much of that whiteboard would look the same and how much would just be totally different? Well I probably if my recollection serves me it was probably a lot of it was trying to compare the on days to cloud and why cloud was such a big deal And I feel like my predictive capabilities were pretty locked in you know sort of maybe short of AGI But the whole idea was software is going to move from your data centers, it's going to move to the internet. And in the process, the real power of that is that it becomes available to way more companies, you know, businesses of all sizes, lines of businesses that never could have used software before, end users. And this was sort of this phase of consumerization of IT. So that kind of played out. And then obviously, we're now in like the next frontier of what is software going to look like in the future. And I think a lot of the kind of core architectural components hold. Like, if you are, you know, running a global, you know, supply chain at a Fortune 500 company, you want deterministic systems and software that power your ERP. If you are at a large B2B sales force, you want to have a clear set of business logic around how your CRM works and how your internal workflows around sales automation work. If you're managing documents for a government agency or a pharma company or a law firm or a large bank, you want to make sure you can secure that data, protect it, govern it, ensure that it's all in a safe place and available to the right people. So all that is staying the same. What has completely changed is probably the interaction patterns on those systems, where the interaction is coming from, and then obviously, like, what you can now do with all that data and information. So the big idea here is that in the future, I would say, you know, probably if today, like, 90% of activity on this software is humans interacting with the interfaces of these tools, probably in three years from now, it would be 90-10 the other direction, which will be agents. interacting with these systems, talking to the data, pulling up data from these tools, interacting with this technology. And 10% might be you going and browsing and looking through the software yourself. Now, the interesting thing, and this is going to be the open debate for the industry, is in that 90-10, did the human side go down by 90% or did we just have a 10x increase in now the agents leveraging these tools? And my argument is probably more in that latter category, which is agents are this explosion of new workers that are all using these systems, which make the technologies even more valuable and useful because you have all of these new workers on these digital platforms that need data, that need to be able to ensure that they're secure, that aren't leaking information in the wrong way. So you need those guardrails still, but now you've got this massive multiplier of what people can do with their data because you have agents that can run in parallel. Right. That makes sense to me. There's this really interesting challenge. Podcast over? Yeah, that's all the time we have, but I really want to thank you for joining us. I think we all learned a lot. No, let's throw in a few more questions for the super fans out there. Because you actually just introduced what I think, you know, it seems like a pretty possibly profound change in the business model for what you all do, right? In these SaaS companies, y'all have gotten used to selling by the seat, right? You have 10 employees, you want 10 of them to be able to use Box, you pay a monthly fee for that. And it seems like that business model is under a lot of pressure in a world where maybe I don't have 10 people in those jobs anymore. I don't actually need the 10 seats. What I need is a business outcome. And you just sort of laid out a case for maybe, well, you know, agents are going to kind of come along and do that. So how are you navigating that? And do you think this seat-based business model survives in SaaS? Yeah. So I think you posited, a scenario that is probably the most open for debate, which is, did the people go away? And in my sort of math that I laid out, the people stayed the same number, but the agents sort of multiplied on top of the platforms. And I think the big question is, there will be some software categories where the literal seats are not as relevant because you don't maybe have as many people doing the work as you did previously. I would actually argue that for a large portion of software categories, that won't be the case. It'll actually be the case that you'll have the same number or more people, but then you'll actually also have 10 times the number of agents as people. And so then it's actually this multiplicative effect of more people or the same number of people or maybe a minor reduction and then vastly more agents. And the part that's not being kind of priced in by the market is sort of is that scenario playing out. And if I just look at a lot of our software consumption internally at Box, there's not a large number of cases that I can make for many of our software products to reduce the number of people that sort of exist as seats. But there's a lot of cases to be made that there's a lot more agentic use cases for that software. So if I take like an objective example that's not Box, if I look at Salesforce as an example, we actually are going to have more sales reps at the end of this year than we had at the start of this year. So that's more seats within the Salesforce universe. And at the same time, I can already imagine, you know, 10 to 100 more agent use cases on the Salesforce platform than I could have two years ago. And those agents might not be, again, roaming around the interface of Salesforce. They will show up inside of Cloud Cowork or inside of Codex or inside of ChatGPT. So the agent will, I will be interacting with the agent via a different interface, but the underlying seat that kind of says, hey, Aaron is a user in this platform. they have this level of access to this type of data, that actually doesn't necessarily go away in this world. And so we're already seeing this within our customers, which is you tend to want a seat for the person because you want some kind of stateful representation of like, what data does that person have? What are their entitlements? What information can they access? But then an agent might do an unbounded level of consumption on the software where I, as a person, can only like click so many things per day, but an agent can go and do that at 100x the scale. So the seat sort of gives me the ability to go in and use my information across these other agents. But then at some scale, you have so much data being used, you have so much access on the platform that then there's a consumption model on top. And this is why I think you're going to actually have this stacking element of the business model in software, which is, you know, humans probably still will have seats, but then agents will be a consumption pattern on top of that. As a CEO, I'm imagining you're looking at all of the SaaS that you guys buy just to run your business. I imagine you might be happy if you didn't have to pay for all of those seats and you could just have agents do it. And you seem very bullish on agents. So when you just look at your own spending on SaaS, your feeling is truly like, I'm happy to keep spending for all the seats that I'm spending on. Well, there's a difference between happy and practical. So, you know, I always would like our IT spend to be less, but I'm also extremely practical about how technology works. And, you know, the phenomenon, as an example, I mean, the challenge is that the bare case of software has actually like, it's like a confusing amalgamation of multiple issues that people have. And it's kind of like a Rorschach test of just like, what do you hate about software? And then you'll see that in the bare case. and so some people are like, well, what we're going to do is we're going to vibe code CRM systems and then other people will say, well, no, we're just not going to have employees and so it'll just be agents and then other people will say, well, we actually just don't need all the features of these SaaS systems so agents will go and do those and some of them I'm sympathetic to, some of them I'm not. The one I'm not, I'm extremely, on that continuum, the one I'm extremely not sympathetic to is we have no projects internally that I know of at least that I've kind of at least approved to kind of vibe code a replacement to an existing SaaS service. Because if I look at the stack of, at least the ones that matter, if I look at the stack of like our ERP system, our HR system, our CRM system, our document management system, it would do us no good to spend our time and energy and IT resources trying to replicate functionality that is already kind of doing its purpose, especially at a moment when I'm about to be able to get 10 times the value from those systems with agents using that data. So if I have to both transition a system that is homegrown and try and figure out the next set of use cases, you actually will just halt your ability to go and innovate and drive any real productivity. And then sort of a minor aside, if you look at the past kind of two to three weeks, I would actually say if you did like a word cloud of the past two to three weeks in AI, one of the biggest word clouds would be cybersecurity. And not the mythos part of cybersecurity, but the like, we leaked customer data, The credentials, the secrets of our system got leaked. We downloaded a package that was exploited. And so now think about if you had the entire economy all trying to rebuild their own version of Salesforce or Workday or an ERP system, and then any one of those events happen, now the entire economy has to halt and do upgrades of all of their systems. Or they have to handle the maintenance and the ongoing kind of improvement of these technologies. So there's sort of like that's just not very logical economically. The part that I actually am sympathetic to is there's some software that as you use agents more and more, some of the value proposition of the software goes more into the agentic layer than it does in the software layer. And in those cases, then obviously you would compress the value proposition of that software. And so at the next renewal cycle, you might not spend as much on that tool. But conversely, and when I think about this as a total pie, I think for every scenario where that happens, there's another scenario where agents will add more value on the system that you're using. And so the net vendor actually has even more leverage in the future. So you might save on one part of the stack, but just end up re-spending it on a different part of the stack because of all the upside of what you can now do. Yeah. I mean, what we're really getting at here is the skepticism that the enterprise software market is facing in this moment. And the reason that I wanted to talk to you first is that to my mind, like Box has been facing this kind of skepticism in various ways its whole existence, right? Like you had to survive this very early pivot from being a consumer company to an enterprise-focused one. You had to convince people, you know, that the cloud would be a safe and profitable place to be. And then you guys just, I think, faced a lot of skepticism about whether Box might just be a feature, you know, rather than a company. And so now you have AI come along, introducing this fresh wave of skepticism into this entire category. And, you know, maybe the sort of most, I don't know, accelerationist version of that argument is like every company is now a feature. And the only thing that matters is going to be the Frontier Labs. So I want to ask you, like, to what extent is this Saspocalypse story just the latest incarnation of a story that has never been true, at least for Box. And to what extent is this AI moment truly something different that is actually going to like rattle the foundations of the business models? Yeah, I mean, so, you know, first of all, I think the market is somewhat parsing the different sort of outcomes a little bit. Not perfectly, but there's a little bit of, you know, kind of discerning behavior happening. So if you look at, for instance, our, you know, if you took Wall Street as one metric of this and looked at our stock, it's held up better than, let's say, like 70%. And one of the reasons for that is that one of the things that is sort of not really under debate is that your most agentic vibe-coded enterprise future still has to store the data somewhere. You still have to go and secure and govern the important information of whatever that workflow is. Like, you can vibe-code the creation of the contract, but the contract still has to get stored somewhere. and it still has to get governed somewhere and it still has to have a retention policy and it still has to have access controls to end users. So we're building a platform that we believe is extremely durable, but the part that I'm excited by is actually becomes meaningfully more important in a world of agents. Because when I think about the use of data in the enterprise, what all these agents really want to do is they want to access data, they want to read data, they want to write data, they want to know context about your organization, they want to know like, what are your best practices what are your policies, what are your customer relationships, what is your research that you're doing? All of that information sits inside of your enterprise data. And most of it sits inside of unstructured data in the form of business content. So we happen to be, you know, firmly on the side of like, bring on all the AI humanly possible, because those agents are going to all be working with enterprise content that still needs to get stored at the end of the day somewhere, it still needs to be secured somewhere. And so the kind of use cases that we're seeing from customers are actually the kind of things that like we sort of only barely could imagine would be possible 10 years ago. Like a customer comes to us and said, we want to automate our entire insurance claim process. Well, what goes into an insurance claim is like a tremendous amount of enterprise content. So when they want to go and do that automation, maybe they build the agent on Anthropic or maybe they build the agent on OpenAI, that agent still needs to talk to all of the data in their enterprise. And so they actually have to go and upgrade their infrastructure to make sure that they have a platform that can work with this information. So I'm very happy about all of this stuff that's happening in the AI space. I think the bigger shift is, and again, there'll be sort of winners and losers in software and SaaS, as it has been true of every kind of era of disruption. So I don't doubt that, yeah, sorry. Tell us about some of the losers. You don't have to name names if you don't want to. I'd rather not. I know you'd rather not, but basically what you're saying, and I believe this, by the way, is like your business just has access to this like very rich, valuable data. And that data is never, or at least not for the foreseeable future, going to be like stored at one of the frontier model companies. So you guys have a lot of value to create just sort of based on that unique advantage. I'm guessing there are other SaaS companies that just don't have that same sort of advantage. So, you know, as you're like scanning across the market, like, is there a business where you're like, well, I don't want to be in that business in a world of agentic AI? You know, I'll give you a framework work to think about. And then I'm not going to name names because I think all businesses right now are in a mode of just like pure execution and people are going to kind of pivot their way through this. But in general, the thing that you want to obviously pivot toward, and some companies are farther ahead of the curve than others on their journey on this, but the factors that you would want to have are, do you have some degree of business logic or workflow in the system? because the agent still needs to do that, even in an agentic world. Do you store data? So are you the natural place for the information to get stored for whatever that workflow is? Do you have a set of domain experience and context about that workflow that the agent, like the next training run of the agent doesn't just replace? So is there some depth in terms of your domain understanding? Is there like an element of like security, governance, trust that matters a lot because that sort of like points to the, you know, how replaceable are you in the stack of, do customers kind of care to replace that? You know, are there network effects of like the more users that use the platform, the better it is? This is sort of one reason why Slack has been just incredibly durable is like, no, like we're already communicating in Slack. So agents naturally show up in Slack as opposed to I'm going to like agentically do Slack. So, you know, network effects, workflows, depth of security compliance, governance, that sort of like, you know, correspond to the level of trust you have to have in the platform. There's probably one more element, which is like, how much does the system benefit from a world of multiple agents needing the data as opposed to one agent needing the data? Because then that would point to whether the enterprise wants to sort of put all of your value into one of the labs or one of the products, or does it need to be a different layer that everything kind of talks to. This is why you still see companies like Databricks or Snowflake are growing quite well. And this is sort of what we're experiencing is actually you don't really want to move your data around constantly. You want your data to be kind of abstracted from where the agent is because you want to be able to structure your data, secure it, govern it, and then let all the agents talk to it. And that sort of reinforces the need for that as a kind of control point. So I would just say there's a continuum. And like, you could probably have like a quotient for like how durable is the platform based on those factors Right The Levy formula Here what I taking away from it If you have a to list app for teams, get out of that business. That's not going to work based on what Aaron just said. I would say that business is actively pivoting. Yeah. Yeah. I actually think I know one that might be. I know Box has talked about AI for a long time, but I'm just curious for you personally, you know, we do have these momentary enthusiasms in Silicon Valley. I think it's fair to say both of us had a crypto phase. I'm imagining that your AI phase started earlier, but maybe sort of took longer to kind of reach the fruition it's at now. But like, did you have a kind of moment of conversion where you like saw a paper, a product, something where you're like, okay, like I need to start taking this really seriously and spinning up teams. Yeah. Well, there's been three moments if I, if I like, and it never is this perfect, but like it almost, you could almost think about it as, as, as like this. Um, there was a moment 10 years ago, more or less, maybe, maybe, maybe nine, eight years ago where like vision models were getting good enough where we saw these scenarios where you, you know, you give, you give the vision model, uh, a document and it could OCR it, or you give a vision model, uh, an image of a retail product and it could kind of classify it properly. That was a pretty big deal because you could imagine that like, okay, like if we could just convert everything into a visual kind of system, then these agents or these AI systems are pretty good at that. The problem on the tech side was you had to train individual models basically for every domain at the time. So this was sort of just before the transformer. And so if you wanted to do like document classification in legal, you had to have a different model than if you wanted to do like financial equity research analysis, which meant that you never had the takeoff moment of AI. So that, but we had the first sense of like, oh, this is a big deal. Obviously, big deal number two, Chachabiti, because while we kind of, you know, we're following GPT-2 and 3, and we had a hackathon where somebody did like GPT-2 inside of a document. And, but it was like, it was like producing like garbled text. It like really wasn't, you know, maybe it could like type ahead, you know, five extra words, but like was not going to game change your productivity. So I didn't have like a complete, you know, religious moment. ChatTBT was sort of the first time in this era where it was like, okay, this is like a very big deal. We're going to be able to now wire up these LLMs and connect to your data and then do meaningful things with them. And then I think probably the most recent one in the past, let's say, year plus has been kind of this kind of, you could probably say it was marked by cloud code. But really these more agentic patterns, which is the LLM runs in a loop. that the agent has access to a set of tools. Those tools might be on your computer. They might be in the cloud. And you can kind of hand off these long-running tasks to these agents and they will just go and do work for you. And then the efficacy of that work has sort of been improving quite a bit. And we've all seen the kind of task length of what an AI can do. And it's been improving at an exponential rate. And so that's like the biggest moment, obviously, in the past 12 months. And now we're just riding that wave. And this might just be the final form factor of AI, which is it's an agent. It could run for a minute or it could run for a month. It has access to any of the data that you need. It has access to all of the tools you work with. It can act as you or it can act as its own entity. But it's just an LLM constantly running in a loop, making decisions. And then you kind of intervene in its work at some point to kind of like steer it or review something. and this appears to be the architecture of the future of AI. Yeah. I mean, I will say, like, I have come to believe that basically everyone should have the experience of building a website using an AI agent of their choice. Like, I feel like a lot of things clicked for me when I started watching a computer use itself. And I think there are, like, a lot of people who haven't had that experience yet and it might be profitable for them, you know, in various ways. You recently posted about how somewhat strangely, this wasn't quite the way you put it, but I'm going to put words in your mouth. AI doesn't seem to be helping any of us work less. Like you mentioned that you'll start working on something with an agent that you think is simple and you lose three hours to it. Was that a real project? And if so, can you share what it was? Yeah. I mean, it's like so pedestrian that I'm embarrassed to share, but I was going into a city and like a week later and I needed to map out a bunch of customers that I should be visiting. And because I had like, you know, half my schedule filled and I had room for some more. So, and this specific example that I think I was referencing, but this happens like three times a week. And this specific example, I was using Perplexity Computer, which does some pretty good workhorse stuff. And I just gave it the task of like rank order all of the top 50 companies in this region, get me the LinkedIn of every single CIO of those companies so I could just make sure I'm like, okay, who have I connected with, whatever. And then I didn't even know what I would do next, but I just wanted to get a good map. And so that task itself took maybe 15 minutes because you prompt it once, you get back some data that you don't really like, you reprompt it, it does better the second time, and then you do it a third time, and then boom, you're off to the races. So maybe like 15 to 30 minutes of AI work. But then the very next thing, you're like, well, I have to do something with this data. So then I spent the next two hours emailing all the people and filling up my calendar more. And it was the kind of task which is like I thought like, oh, I'll use AI to accelerate this thing. And it kind of like worked too well to the point where then I had created more work for myself at the end of it. It would almost have been better if it came back with like full hallucinations because then I could have just gone to bed and be like, well, that failed. But it like worked. And like by 10 p.m., I was like, well, now I'm going to feel bad if I don't actually like do anything with the tokens I just exhausted. And so it's a very small, you know, metaphor or anecdote that I think is happening everywhere, which is like you're like, oh, I'm going to go tell AI to write this like little web app that I want. And then you're like, well, well, I built it. Now I've got to like, oh, I want to add this other feature. And I should probably like get it hosted somewhere. And then, oh, I might need to like go and change this one thing. And it's like, you know, it's basically if you give a mouse a cookie applied to the economy. 100%. Like you just start building up more and more work. Now, there's a little bit of like multiple reasons this is happening in Silicon Valley specifically. And some of those reasons will not correspond to the rest of the world. Some of it is like a Valley-specific thing, which is like we are just drinking from the fire hose of this new technology. but but the part that that i find really interesting and and this is it tangentially relates to why i think the job loss argument is wrong i think that the part that is um that that will translate is people will find that that basically like there was way more tasks that they could be doing that they just never could do before because the the the sort of fixed cost of starting the task was too high but ai made it ai made it easy to get going with that they lit up the project. They did the research. They did the analysis. They reached out to the customer. And that then kicks off a cycle of downstream work or a new set of constraints that start to emerge. And so my hunch is actually, as we start to use these things as these sort of abundance intelligence tools, we're just going to find the next bottleneck in our company or our firm or our team. And then we're going to say, well, was that thing valuable enough that I actually want to see it through? And if it is, that will probably then still eventually hop over to a human needing to do something, in which case that is the new type of work that gets produced. And so, and so just, just, just to bring this all the way home and then, and then, and then, and then whatever you want to jump in on the task worked actually so well that like, I idly wonder, should I have a full-time person just doing this task with an agent? Like they will use an agent, but like, I don't want to do this every night for the rest of my life. I like, I, between nine, 9 PM and midnight, I don't want to be doing this, but, but it might be valuable enough that it's worth a person to do this for me, in which case it actually created a job because of my experimentation of this AI that I was just doing for fun. Yes. I mean, so I love that story and I feel like I have lived both sort of sides of it. When I built my own website, it was absolutely give a mouse a cookie because of course, after the website was done, I said, well, I need to host it. And then I said, well, it should probably have a blog. So I added a blog and then I said, hey, when I log on to the Why isn't it telling me the current weather in San Francisco? Like I have to solve this problem for the three annual visitors to this website, right? So I sort of went down that path. Now, it was super fun. I don't regret the time. I also didn't, I think, create a lot of value for myself. On the flip side, though, you know, I run what I sometimes think of as a somewhat fake business in the sense that people pay me to email them. So it's just a very sort of strange, uncomplicated business. and I have a bookkeeper and I have an accountant and they send me a monthly email sort of letting things, letting me know how things are going. But I've never really done like real kind of financial analysis and then Claude Cowork shows up and I can just start chucking spreadsheets into it. And for the first time since I started my company five years ago, I'm like, tell me about my business. Like, what are my problems? Oh, actually, that's our revenue. Oh, wow. Yeah, yeah, completely. And it starts to feel much more like that latter thing, which is like, wait, this is something I never would have done before, but it actually is valuable. And it pushed me to do, it pushed me to make some changes, including by the way, starting this series of conversations that you were the first person on. So my question is like, what advice do you give for people when, when they're like, okay, I'm bought in, I'm going to try this stuff, but how do I know when I'm just spinning my wheels versus like creating something of durable value in my company? This one's hard because it's hard to answer generically. But, but I would, I would say this, will be like a defining question of the next decade, which is if you have access to abundant intelligence, but it's not free, and so it's abundant but not free, how do you allocate the sort of spend of these tokens and the use of these tokens across the organization? And I think it's like on one hand, it's easier the higher up in the organization you go because you have all the data about what the company does well and what it doesn't do well. And on the other hand, it's actually somewhat easier, lower in the organization where the work is actually happening because you can kind of like – you can kind of self-identify the work that matters or doesn't. But the problem is both of those things have sort of problems, which is the direct user of the tool might think that their work is the most important for the tokens as well as the person high up the organization might not know about the new innovation that somebody has. So you've got this crazy new problem, which was to some extent, the world was a little bit easier with with with scarce intelligence because you it was everything was kind of slow and it had to be slow. Now, abundance and abundant intelligence, you could have everybody running around, you know, spinning up agents, doing lots of work that maybe 70 percent of which is like not valuable. But you don't know which is the 30, 70 until you've done the whole the whole set of things. So I think companies right now are trying to figure this out. It's a it's a it's a brand new problem. I almost don't have any great answers because each company will be slightly different. I think directionally, the rough shape of things to me would look something like most companies, if they've survived 10 years in, generally know the shape of where their company makes money and where they don't make money. And I mean that, I know it sounds incredibly lame, but a pharma company understands that like the more research they do in a particular path, the more accelerated a new discovery might happen. The more they can accelerate a drug trial process, the faster they can get the drug to market, the better they can go in and get these products distributed to all of their layers of distribution, the more sales they'll get. So where would you apply tokens at a life sciences company? You'd apply tokens in those areas. What you probably wouldn't worry about is like, like, you know, expense reports. And like that probably wouldn't be the place to go because like that's not the constraint on on your ability to grow faster as an organization. Now, you might still automate that, whatever. But like, that's not where the the corporate energy is going to go and apply this this, you know, expensive compute resource across the organization. So so I think industries kind of know that life sciences knows it's about discovery and development. Banking knows is about, you know, private wealth management, investment banking and trading. manufacturing is sort of what's the throughput or cycle time of your supply chain. And so the tokens will basically flow to those areas. And then the only question is as a user, as an end user trying to get more productivity, what's the one click down that sort of matters? And this is where, you know, you have to do a little bit of like iteration with these tools to figure out like, like, okay, organizing my desktop wasn't that useful. It didn't like change my productivity and it was expensive versus like doing advanced life sciences research on this data set was expensive, but very useful. And so you kind of want to like apply the compute in those areas. That makes sense. Is one potential solution to just put up a leaderboard of who has burned the most tokens in a given month and reward them somehow? Okay, good. So that is emerging as the best practice. So the alternative is just token max and you're good. But actually, but like, But, you know, token maxing aside, I would say that there's a tool set that I'm sure like 10 startups are working on right now. So they're going to see this and then they're going to all pitch us. But it'll be a good business for a few of them, which is there's probably a new kind of like ERP HR finance system that lets you have a heat map of where the tokens are going and then the rough value allocation of what that produced. because right now it's kind of a comical idea of like, oh, you're going to have to treat tokens like headcount. And you only put headcount in the most important areas, but it very clearly is going to be the same case. We are only going to be able to apply compute to rationally the most important areas of the business. And so we do need some visibility into that. And so there's almost like, what's the performance management system of tokens is like a thing that has to just get built for companies because right now it might be 1% of your company spend. But if in three years from now, it's 10% of the spend, we don't joke around with where 10% of our labor force is going to go. We're incredibly maniacal about how do we apply that. The same will be true of your tokens. Right. Let's talk about jobs. You recently posted about a kind of Gell-Man amnesia for AI. Gell-Man, of course, is where you read about something you know a lot about and notice the obvious errors, but assume that that same source is credible on other topics. And so you wrote, quote, people use AI for their job and see all the various things they have to do in the last mile, but then look at someone else's job and think that AI will eliminate it immediately. This is absolutely true, by the way. Why are some people so quick to think that AI can automate away a whole job? I think we, first of all, this is amazing technology. And like, it is the coolest technology, you know, at least that I've ever played with in my life. Um, and so to some extent that's like, it's like deceptively cool because, because it's like, like, Oh my God, I think I just did my, my, my taxes or like, Oh my God, I just built this amazing marketing website in like five minutes. And, and then we, we sort of like look at the output and we're like, gosh, that must like totally replace the job of like X, Y, Z profession. And there's a, there's a few asks, there's a few kind of like core flaws with that, which is like, well, what is that profession, you know, doing for, for all the hours in their day. And like, how much of it is, is, you know, just doing the final calculation of your taxes versus no, it's like getting all of your data in order. It's reviewing all of the work in the process. It's asking you questions back and forth. I'm like, what are you optimizing for? And it's knowing the right questions to ask. It's, it's like, you know, dealing with the, the three missing things that, that you didn't even remember that you're supposed to add. But, but if like an AI system had done it, it would have totally glossed over. Like, that's what the profession does. And the automation of one or two or five of the steps are just these individual tasks that it's able to automate. In the case of development like we seeing this day in and day out is like you or I can go and tell CloudCode generate me the XYZ product And we could be like wow that must automate the engineer out of existence Well, there's a couple issues. Like one, like the code quality is probably horrendous. Like the ability to like now ask it to do 40 other things over a 12-month period is just going to stack in complexity. the moment that there is you want to actually get that that software hosted and run and make sure there's no downtime and ensure that you have a good distributed system is already 100 times more complex than you just prompting the the code to get written the moment that there's a security event all of a sudden somebody's got to like wake up and respond to that um you know i can name 30 other things that a developer has to do like you have to like understand like what's the roadmap where's the company going all of those things and so all of a sudden you're like oh the job of the engineer is absolutely writing code. And there's a lot of people that say like, you know, the job of the engineer was never to write code. It was to do X, but it's like, no, no, they're like, they're writing code most of the time in the, in the prior, you know, world of work. The problem is, is that you were highly constrained by how much code they could write in a day. And then, and then they were just automatically bottlenecked by that to do the other things that their job could be. And so what is the future engineer? It's like, yes, it's to understand the, the, what are you trying to build? It's to make sure that it gets built properly. It's to ensure that there's no security issues. It's ensure that it gets released. It's ensure that it's high quality, all of those things. And so if you or I go and vibe code something, we think we've replaced the engineer. We think we've replaced the accountant. We think we've replaced the lawyer that we get advice from. But then you actually go and look at like, okay, that was like, that was like, you know, that was the first 80% of the job. But the extra 20%, it turns out is like all of the value creation of that profession goes into that last 20%. And all of the expertise and domain knowledge is in that last 20%, not the, not the text that got generated. Um, and, uh, and so, and so that's like the misunderstanding. And then, and then the converse is like, we look at AI and we're like, we're like, we want to use it for X, Y, Z thing. I'll use it for like analysis of like a market that I'm thinking about. And if I just took the output of that analysis and I ran with it, I know it would not work because I know that it's missing context that, that, that it doesn't know because either I didn't give it or I know something else about a different trend, but somebody else might see that and say, wow, Aaron's job is like incredibly easy. Um, and like AI just gave me the answer of what he's going to go do. And, and then I'm like, I'm like, no, actually my job is way harder. I promise. Um, and I think that's kind of what, what is sort of happening is like the moment you have a five or 10% flaw in what the answer was, the moment you have a security event, the moment you have to actually like see the whole thing through, that's really what you're paying that worker to go and do. Um, and, uh, and I think you're, you're just, you're, we're going to have this duality of like, we're going to see these amazing things and then you're gonna have to go in and, and see, and then that last mile of real work that has to happen, it doesn't go away. And, and, and it just, the last mile keeps moving and moving. And so, you know, um, uh, you know, you look at, you know, Dario has this interesting thing about, about, about kind of like, you know, you first automate the 90% and then you automate the next thing. And then, and then, and then you kind of increasingly sort of like are automating out the software development life cycle. But I would almost, I would, I would say there's a different axis that, that, that maybe, you know, people need to think about, which is if you took today's static work, that actually maybe would work. It's like, you'll get the first 90% and then we're going to automate the next 9% and then we're going to automate the next 0.9% and so on. But actually what happens is, is there's a dynamic part of the equation that's not represented by that, which is the market is starting to ask more from the provider because they now know what is possible. And so just as you automated that first 90%, all of a sudden the market shifted on you. And that 90% is now like the new 50%. Because actually the demands of what you ask an engineer to do just go up tenfold because you're Like, I think you can do that thing way faster now. So I'm going to give you a much bigger project. Or I think that my, you know, tax accountant or my lawyer should do way bigger analysis of the topic at hand as opposed to just giving me like a rudimentary answer. So you have this other dynamic system that's happening, which is our needs and demands are just growing as a result of what we can go and automate. I mean, I love the idea that AI will just sort of let us re-envision what our jobs could be in a sort of more expansive, creative vision, right? My fear is that the last mile of super, my fear is that the last mile of human supervision will just turn out to be like kind of boring, right? It's like, I feel like we're already starting to see this in some jobs where it's like, there is a piece of this that is automated and my job is now just to review AI output. and that is just like pure drudgery. So is that a factor here? And does that like complicate the picture that you just painted? Well, it doesn't complicate the picture. It adds a, because the big question is, are there jobs in the future? And the answer is yes. Now the question is, do we want those jobs? Like of reviewing the output of AI agents. So maybe we all just like opt out of the economy because we're like, God, I don't want that job. But so interesting philosophical question of like, what is the new way to get fulfillment and creativity out of these jobs? And you can see it. You see burnout of engineers on Twitter or X that are basically like, all my job is to review, you know, slop from the AI. And like, that's, there's a limit to how fun that is. And so I, you know, I think super interesting question. I think there will be a continuum of jobs though, as an example, because, you know, engineers are facing this first and they're also facing the existential kind of dread first, but engineering is, is a very unique job compared to the rest of the economy. And we can kind of get in some of the differences and actually why I think diffusion will take longer than most people think when they just look at the engineering work. But engineering work is like, you know, most of your day is like, like, I'm not, I'm not trying to be reductive, but like your, your job is to, you know, like obviously think about a problem, think about a system, write code, and that code is text. So you're just write a lot of text and then somebody else reviews the text and you ship it. And so if the agent just did the writing and review writing of the text, then all you really are doing is reviewing the text and then shipping it. And that's, there is a, there's a new form of like, what is the job of the engineer? It's more about the system you're building. It's more about the customer problem. That's a journey that engineering is going to go through, but you go into like, go talk to an investment banker, go talk to a lawyer, you know, above sort of paralegal, go talk to a doctor or, you know, a, you know, kind of a nurse or go talk to a, um, uh, a pharma researcher. Like they would love to it out of the toil of, of like, like, yeah, I actually don't want to spend 15 hours generating a corporate pitch deck for this client that is just me moving around some, some images on a PowerPoint presentation, doing some Google searches to find market trends and then pasting that in. I'd love to automate that. And then the job that I should be doing is I should be getting in front of my clients and I should be spending time making sure that we're like delivering unique value and insights to them. So, so that's why I'm not like overly worried about this because I think that, that we're seeing some hyper accelerated dynamics in engineering that don't always relate to other, other forms of work. And like, if you talk to a doctor and you say like, how, how much do you enjoy like typing up the patient notes after the, the, the, the meeting of the, the, the patient meeting, they want to automate the heck out of that. They want to be done with that part of the job. So I think you're going to see like, mostly it'll be a net positive change as people are able to get rid of the stuff that they hate doing, start to do more of the stuff they want to do, and the sort of demands of the job evolve in such a way that makes this stuff just much more exciting because it ups the level of the kind of work that you're producing. Yeah. It's the sort of thing that I love to hear. I would love to not live through a massive disruption where we see super high unemployment. I also can't help but note we saw almost 46,000 tech layoffs announced in March alone, with AI sometimes being cited as a potential cause. So I'm curious if you would put a number to it on the software engineering front. Do you think in three years we have about as many software engineers we have today or more? Or do you think that there is a bigger shift there? I think we're going to have more. Um, and, um, and the, the thing that, that is, um, that the thing that, that I think is happening that is, is kind of important to, to Parson, and this is, this is, uh, there will be disruption. I don't want to be, um, you know, uh, uns, uns, unsympathetic to, to, you know, people that really will, will fear, you know, face these kinds of changes, but I, but, but let me just do like big picture for a second. Um, if you were, you know, a CS grad, uh, of the past two decades, of a top 25, 50 CS school, you basically, by and large, maybe you went into a banker consulting or something, but by and large, you were trying to go to a tech company. And by and large, that tech company was in Silicon Valley or a couple other places. And so most of the software talent in the world, at least of this kind of cohort, ended up building software for consumers. And then we were building apps and building ride sharing and building enterprise software, thank God. And that was like all the stuff that we were building. And we've accumulated a lot of engineers on that kind of work. And some of those companies have overhired and many, many dynamics on that front. The thing that's about to happen, and especially, so who was the loser of that equation? The loser of that equation was every other company on the planet because they couldn't compete with Google and Facebook and Microsoft and so on for that top engineer. And so they couldn't then go and automate XYZ thing in the life sciences process or in the supply chain or in the automotive kind of AI system. I don't know how much software you've used from companies that aren't in the Valley, but if you log into your bank and you're happy, you're like a totally rare person. And if you look at most average kind of car console designs of any car that's not like two companies, and you imagine how, you know, unusable these systems are, that correlates to the fact that they were not able to like overstaff with all the top engineers and designers and all the people. So now what happens? All of a sudden, those companies, what maybe was a 30 or 50 engineer problem previously, now Cloud Codex comes in, and all of a sudden it's a 5 or 10 engineer problem. and for the first time ever, they're able to go and take on the work that wasn't possible before. And they're able to go and bring automation to all of the systems, all of the workflows, all of the products that they're building that they could not have afforded or budgeted or been able to justify automation for. So what's going to happen is, in some cases of tech, you'll see a temporary dislocation. At the exact same time, the thing you should be tracking is the number of engineering jobs that are opening up at traditionally non-Silicon Valley tech companies. And what does that start to look like? And they could be small businesses, they could be consulting firms, they could be life sciences, they could be manufacturing. And then at the exact same time, just as a force multiplier to all of this, because of AI entering the workplace, you're going to have a number of new types of engineering jobs where that job is entirely about how do you deploy agents inside the firm that can go and automate work. And just as an example, I did this for fun just to make sure I wasn't kind of full of shit. If you go to the Eli Lilly careers page, as one does, and you look up, you look up the careers that they have, they have this job title called lab automation software advisor. That person is an engineer whose job it is, is to go and bring automation through AI to the lab process. Think about how many hundreds of thousands or millions of jobs will look like that in the future, which is my job is to take the innovation coming from AI land and apply it to this particular business process in my organization. And you're kind of like an FDE, but for that company. And those will be the people that would have gone to Meta or would have gone to Google five years ago. They are just going to now work in pharma. They're going to work in banking. They're going to work in manufacturing. And oh, by the way, those are actually incredibly stimulating jobs. You're just not building an app. You're automating drug discovery. So that's the shift that we just sort of have to go through. And it's going to be a little bit messy at times. And we don't have as many of the stories of a company hired 10,000 people because it's easier to lay off 10,000 than sort of, you know, seeing how that distribution ends up across the rest of the companies. But that is the evolution that's going to happen. So you're saying that even in the future, like even just with my small newsletter business, I might one day be able to fulfill my lifelong dream, which is to compete directly with Palantir and just sort of like build like a vast sort of like surveillance and analysis program. If you so choose, you can. But these are the kind of jobs that are going to be opening up is all of these companies are going to need people to go and implement this stuff. What is the fastest growing role probably at something like Anthropic or OpenAI? It's these FDE roles because you actually need humans. The forward deployed engineers. Yeah, you need humans to go and implement this stuff inside the organization. Those are just what, and those are the engineers of the future if you're not just building software that is an application to go use. So that's only one of the many types of jobs that I think opens up. Like that's the most sort of like easy to understand because AI, you know, it has to be implemented. But I think there's a lot of jobs. As an example, you know, I don't know if you saw – I'm sure – do you see that Dwarkesh was trying to hire like a person to go help research and, you know, do whatever. I missed that. This is Dwarkesh Patel, the great Silicon Valley podcaster. But I missed this. So he's looking for a researcher. But this is like nine months ago or something like that. Okay. And it wasn't even done ironically. It was just like, yeah, I have a job opening. I was like, well, why is Darquesh the most AGI-pilled person on the planet? Why does he need to hire a person to go and help him with his podcast? It's because it just turns out that we're just going to do more work because there's just way more stuff to process. There's way more information to go and consume. There's way more workflows to get involved in. There's a funny article out of the FT, And, you know, I don't know how much I trust their views on technology, but it was that lawyers are being inundated by clients that are asking them questions because they went to AI. And they have to, like, verify that the advice is good or they have to go and review the sort of contract that got written. So what we do is we're lowering the barrier for everybody to participate in these things in kind of like a touristy way. Like I can, like, be a tourist in software development. I can be a tourist in legal. I can be a tourist in health care. but that eventually still needs to get verified or the work actually has to get done that last mile. And, and that eventually still moves into then you need some kind of like semi expert to go and do this, which is why I actually don't know that you can yet claim what, what, what, um, what, uh, degree should you go into in college? I don't think we yet know that, that one of the degrees are off the table. Like, I think you should totally go into CS if you're really excited about software development. You just shouldn't expect to go build a little app that you press a button on. You should expect that you're going to use CS skills to go and do clinical trial automation at a pharma company. Well, it's, I think, a great place to land because it leaves me with a feeling that I have so rarely when thinking about the tech-enabled future lately, which is optimism. So thank you, Aaron, for giving us a jolt of that. and would be great to maybe check in with you again in a year and see if the picture is still as rosy. Awesome. Aaron, thanks so much for joining us. Thanks, man. Platformer is produced by Lindsay Chu and edited by Fitzharris at Story & Sound. You can watch this whole episode on YouTube at youtube.com slash Casey Newton. My email is casey at platformer.news and we'll see you next week. Take your team from AI novice to AI native with Atlassian Rovo. Go to rovo.com to learn more today.