Why the "SaaS is dead" narrative is completely wrong - Aaron Levie [box]

Speaker A: AI is too important for the rest of the world to say. We're only going to use the API calls from three companies in the us. There's a bunch of people secretly cheering that this is exactly what you'd want to have happen to design the regulatory review process for AI model releases. Intelligence is going to be multifaceted. It's not going to be one or two models doing everything.

Speaker B: Today on Billions, I'm sitting down with Aaron Levy. He's the leader who rapidly transformed BOX from a pure SaaS business into a cutting edge platform for AI agents. We dive into the high stakes battle for sovereign AI government regulations and why the narrative that SAS is dead might be completely wrong. Aaron, super excited to have you on. Super excited to uh, have you on. Like uh, I know you are like you have very interesting take about uh, AI. Uh, I'm curious to like get uh, your uh, last take on what happened with uh, Anthropic recently. Uh, because I think you said it was like a big win for uh, open and lightweight models. So can you maybe uh, elaborate and uh, your thoughts on the topic?

Speaker A: Yeah, um, so I think first of all the Fable model is um, kind of an unprecedented level of intelligence and capability. Um, uh, so clearly Anthropic designed an incredible and trained an incredible model. Uh, but the US government uh, effectively enacting an export control which meant that non US citizens aren't able to use the model is uh, really kind of a completely new precedent and moment in AI regulation where we've never seen that before, uh, for any kind of frontier model being held back legally uh, from being available in the market. And the implications are um, first of all it's almost entirely impractical for any AI, an AI lab, uh, to be fully responsible for the nationality of the customers that are using it. At a minimum it would completely obliterate the API business of Anthropic because the idea that you would be able to control the end customer through your partner ecosystem is very, very difficult. So it effectively has the impact of making the model unavailable to the world. Uh, and so the risk is if you were uh, either a major app developer, um, and you want to be able to make this technology available to your custom customers, or uh, you're a country anywhere around the world that wants to continue to ensure that uh, your ecosystem is able to build on top of leading frontier AI, this poses an all new risk which is at any moment can that uh, AI model get pulled back uh, from one government, uh, or due to one government and we just never had that precedent before. AI has been this very fast moving, very market based, uh, ecosystem over the past couple of years. Uh and so you know, I think first of all the US Government probably very quickly needs to uh, to improve the situation with Anthropic. I think Anthropic is quite motivated to you know, resolve whatever the, the particular issues are. Uh, but even, even when that happens, it's easy to kind of think in a year from now or three years from now or five years from now, how many more versions of this might we see, uh, in uh, in, in, in this landscape, uh, that that will play out. And so it sort of makes you wonder. Well, if you're you know a country, Germany, uh, France, Japan, um, you know, uh, uh, UK and you have the capability to, to you know, be it at you know, frontier or near frontier AI, uh, from a talent and an infrastructure standpoint, do you need to be, you know, you know, building these uh, these kinds of models yourself and make sure that you have them available to you know, businesses and users within your country? I think it, it really kind of introduces that question bigger way. Um, and we've already seen obviously examples of this, China obviously developing you know, kind of great open source models, Mistral, uh, you know, doing the same. Uh, but I think this highlights the increased importance in that dynamic and the idea that you can have sovereign AI or AI that you control. Now there's one, there's sort of one caveat to all of this though, which is, which is um, uh, you know, you're not going to, you're not going to take that big of a, of a sacrifice or a discount on your, on the model's intelligence, uh, for that benefit. So if Frontier from us, uh, in a, in a kind of a closed model environment stays 10, 20, 30, 50% better than the next best model that you could, you know, make sovereign, uh, I think you know, people will take that risk and they will use the frontier model at all times. So this only works if open uh, weights models or you know, sovereign AI can remain close to the frontier, um, uh, and much closer than it has historically. And I think we're seeing some examples with, with Kimi and Deepseek and um, and uh, and a few of these new models coming out of China and Nemo Tron. It's, it's. These models are doing a great job kind of staying at the forefront, but they're still not fable level. Um and that's going to be the big question for the ecosystem over the, over the coming kind of months. And years.

Speaker B: Do you feel like it's only going to be a matter of uh, performance like staying, let's say as you said, 20, 30%, uh behind or also a matter of timing? Because what we're seeing right now is like when the first model went out, uh, the first lab to catch up maybe took like uh, 12 months or uh, uh, a bit more than a year. But now I feel like models are catching up within uh, a couple of months. Uh, so do you think it's only a matter of difference or also a matter of how fast can other models catch up on the best models?

Speaker A: Well, I think um, uh, you know there's, there's a big debate, um, uh, you know, and if you watch, you know, Dwarkash, talk to Dario or Jensen, you know, you sort of see examples and contours of this debate, which is uh, is the frontier, uh, and the open and the sort of, let's say the best in class open weights model. How close are these tracking versus Are we seeing a divergence? Are they six months behind or are they three months behind? Are they one month behind or how much Is it really just a factor of how much compute you have, uh, for inference and kind of test time compute as an example? Because there's different ways to, to kind of really drive the workloads in these models to get better results. Um, I am um, generally actually optimistic that uh, that um, uh, that the frontier open weights models will be able to, to stay relatively um, caught up. I think it's, it's of such uh, economic and uh, strategic importance for you know, countries like China. Uh, you know, you know, maybe individual companies like, like Mistral or even Nvidia to make sure that, that, that, that is sort of maintained. Um, and we have not yet seen a breakthrough that is so either proprietary or kind of undiscovered by the rest of the ecosystem that keeps this completely closed off. Um, you know, I think there's obviously a conversation around um, you know, RSI and, and, and what, um, uh, you know, what, what does self improvement look like? And does that kind of lead to a kind of um, an accelerating takeoff like whoever, whoever you know, is able to design these systems that, that self improve first. Then obviously there's a compounding dynamic to that. Um, at the same time it's hard to imagine China not wanting to stay in the race in a very big way. Like it's unlikely that China says, you know what, we're totally fine with anthropic@OpenAI being the only vendors of AI in the world, uh, and we'll be closed off from this ecosystem. So at some point uh, you can just throw enough money at this problem. China could just say we're going to spend $50 billion. Um, and just, and just make sure that this is of such critical national security importance that we have to make sure that we're throwing compute at this problem. We're going to kind of, you know, sequester the top, you know, researchers and scientists, um, and uh, make it happen. I think it's unlikely that that wouldn't work. Um, and I haven't seen any evidence that suggests otherwise.

Speaker B: So your view like ah, when it comes to China uh, is what regarding like the chips? Because for so long uh, you know there's been kind of uh, an embargo on chips and it was not possible for them to get them and they were getting it through like Singapore or like all these kind of things. Like what, what's kind of your view? Because it's like, it feels like eventually like I uh, mean in the US it has always been very pro open market and now it feels like things are moving away from it. So I'm curious to get your take. Like.

Speaker A: Well, if you uh, and I think a lot of very smart people believe this, if you, if you believe that AI is uh, is both a national security, uh, you know, threat and opportunity and it's a, you know, of, of critical strategic economic imperative that can lead some, some people down the path of. Well then you, you actually want this to be a closed environment. You need to keep the chips, you need to keep the models. Um, and uh, and I do think it's of national security, you know, level of, of importance. But there's this interesting tension which is uh, which is if you're also not the exporter of the technology, uh, then the ecosystem effects start to favor the non US companies. And Jensen uh, made this case in the Dorcash podcast about a month ago. Uh, and actually I probably lean more toward uh, Jensen's side of this argument. Which is, which is you know, if you just play this out, AI is too important for you know, the rest of the world to say, you know, oh yeah, we're only going to use the API calls From you know, three companies in the U.S. it's, it's just too critical of a uh, of a, of a technology. So sovereign clouds have to get built out. Uh, GPU clusters are going to get built out globally. You know, countries are going to want to train their own models like that. That's an inevitability so if that's an inevitability, then you have a choice. Do you want your technology stack, do you want your architecture to be the architecture that powers all that, or do you want to force the rest of the world to have to go and design their other architecture? And if that happens, who stands the game the most? Obviously China. Because they will just eventually stand up enough, enough compute, enough infrastructure and make that happen. Now I think this is a very um, worthwhile, uh, debate and I am only like, like 80% confident in my opinion. Uh, the 20% where I'm not confident is you could, you could possibly squint and argue that America is just so far ahead in compute and chip design. And now we're starting to kind of fig how to do fabs, um, and you've got Elon with Terrafab and we know how to do the orbital data centers. Maybe in the future that maybe there's one scenario that you play out and that is that we do actually just have the API for intelligence, um, and that we control the access to it, um, and we can make sure that only our allies are using it in the right ways and we can prevent the bad actors. I think there's, there's certainly a scenario that looks like that. That's what many of the kind of AI safety, kind uh, of ecosystem is sort of hoping for, um, as the outcome. And I totally grant there's some percentage possibility of that. Um, but I kind of default to this is such critical infrastructure and so important that it will cause other countries to have to just go their own direction if the U.S. uh, uh, isn't enabling that.

Speaker B: Yeah, personally I'm more on the side of uh, people should let uh, Jensen win. He has the chip, should sell them everywhere.

Speaker A: What's interesting is you can already now see the downstream impact. So Jensen was the first to experience this, uh, but anthropic is now the next to experience. It's the same exact thing. So uh, if you already believe in the precedent that you shouldn't export a chip to China or some other country designing kind of leading AI, then it stands to reason you probably also aren't going to let the leading AI model that could help train the next model or help crack systems be available to those countries, uh, or certain ecosystems. So we now have, because of the chip export controls, you have a lot of the foundation for now what we're going to be dealing with. And again the risk is that creates a very unstable environment, uh, for countries globally. Um, now again, as long as we have the best models. You know, maybe you just, you can just, you know, power through that, but it creates a lot of incentive for other countries to, uh, to want to respond in kind.

Speaker B: Yeah. And, and do you feel like, uh, looking at Anthropic, like, marketing in the recent months, I mean, it's been pretty common for models to make this huge announcement that their next model shouldn't be released because it's too dangerous. And we, we saw that, you know, with, uh, OpenAI. So obviously what it does is like, everyone gets excited about the model. Yeah, Anthropic did that six, uh, months ago, and now they've done it again with Mitos and Fable. Like, do you feel like, uh, their marketing went a bit too far and it's just like, uh, going against them, or do you really feel.

Speaker A: Yeah, yeah, Sorry.

Speaker B: Sorry.

Speaker A: No, no, no.

Speaker B: I was just wondering because, like, uh, David Sacks, like, uh, was sometimes like, working at, uh, with uh, the White House. Like, what he was saying is that essentially like, they emailed the Anthropic team and they told us like, hey, can you please, uh, deploy your patch on that specific model? And apparently the Entropy team hasn't been like, that responsive. So we're hearing like, kind of two side of the story. And I think, like, in the marketing of Entropic, it's been very real from the start that they want to be seen as the good guys versus evil Samatman, which I think is not super fair when you look at the global environment. So, yeah, I would love to get your take on this.

Speaker A: And uh, yeah, I, you know, this is, uh, this is one of the more complicated topics right now because, um. So let's start with the easiest part. Uh, unquestionably the rhetoric that has come out of Anthropic, uh, and other, you know, kind of safety researchers created the environment that we're now in where if you kind of scare the crap out of the government, uh, by saying, this is, you know, the most dangerous thing we've ever seen, it has all these risks, and then the government sort of, you know, you know, we can debate how. How kind of compelling the. The particular jailbreak is, but if they eventually discover that there's a jailbreak on what you claim is now the safe version, and then it. It defaults back to the unsafe version, or it can do something that it. It perceives as unsafe, well, that's. That's clearly a byproduct of. Of creating the atmosphere where everybody is concerned about this. So like, unquestionably it's coming from that tone and that, that kind of, of, of sort of um, uh, you know, kind of rhetoric. So that's the first layer. So, so like that, that's I think quite, quite straightforward. It's pretty obvious. Uh, and, and so we're kind of dealing with now the, the uh, the consequence of that. The interesting thing that I, what I don't know though is you know, is it, is it plausible that there's a scenario here where actually Anthropic prefers, weirdly prefers this outcome, which is. Now we have a precedent that has been set that models at a certain threshold of capability do have to go through an approval process, they do have to go through a review process, they do have to be approved and vetted uh, by the government. That's not so different from what the regulatory lobbying uh, has been pushing for from you know, many of the AI safety researchers. And so, so interestingly, if you kind of study the journey of this whole space, the top AI researchers in safety, uh, and kind of you know, regulations and controls probably would have designed a system that says the government is going to approve the release of models at a certain capability level. We're going to do lots of red teaming, we're going to kind of get in a room and we're going to agree that the thing is safe to release. And nobody had kind of created the catalyst previously to just build that environment. So I think there's one other layer which I hold out a little bit of percentage of. Uh, my thought process on is maybe this is exactly what Anthropic wanted to happen. We're all dealing with the day to day drama of wow, I can't believe they responded that way. Or I can't believe Amazon is the one. You know, we're caught up in the, in the sort of all the little leaks and the messages when actually like you zoom out and you're like this is sort of how you would design the regulatory process of AI model releases. If you were a deep AI sort of safety research oriented person that was, that really was concerned about these kinds of risks. Now I happen to not be in that camp. So I think this is sort of a bad outcome. But I can, I could totally understand, I could totally be convinced very easily that there's a bunch of people secretly cheering that this is exactly what you'd want to have happen to design the regulatory review process for AI model releases.

Speaker B: Yeah, makes sense. And um, when you look at the competitors, uh, talking about, I think for open source models, uh, in China there is always a question regarding the data and where it goes and uh, how it's used, etc. And uh, when you look at companies like Mistral who have positioned themselves, you know, really towards uh, more enterprise and helping enterprise like build on top of their models recently we've seen, I uh, think Cursor is actually like a uh, really good use case of how you can specialize a model towards something very specific and get massive value on top of it. So how do you see this play out and do you feel like uh, like Mistral is potentially the one that is uh, positioned the best toward this because they've already uh, a deep, you know, roots inside enterprise and governments etc. To develop on top of their uh, of their own data?

Speaker A: Yeah, I really like uh, first of all I really like Mistral's kind of market uh, opportunity m much more today than maybe three years ago. It's actually interesting as AI gets more important, as the capabilities get more capable, more effective, as agentic workloads cost more money, um, all of those things start to favor uh, a bit more of a uh, uh, kind of multiplayer ecosystem as opposed to sort of just one model from one or two players. So I think Mistral is a kind of a net winner in this. I think again one or two of the kind of Chinese labs are. I don't know if they're financially net winners, but they're kind of net winners on using those models. There's these companies, Base 10 and Fireworks. They're net winners because you're going to go and do RL on uh, on one of these foundation models um, that are open and then, and then kind of you know, sort of post train it for your particular use case. I think that's a winning strategy. We saw that from Cursor. We're seeing this more and more in a lot of the kind of applied AI companies. So I think the big update of the past, frankly month to two months and then the past four days is, is actually like intelligence is going to be multifaceted. It's not going to be one or two models doing everything. And uh, there's even winners really in the applied layer. Like Cursor is kind of this great case study now of we are going to build a harness. It's going to get really good at one thing which is coding. That harness can route workloads to different models. And oh by the way, we're going to have our own cheap model that's fast. We're also going to let you use Fable and let you use codecs and um, gbd55. And then, you know, sort of. I think that's kind of a pretty good stable equilibrium in all of this. And then everybody has to compete on some dimension of okay, my model is better at this use case, my model's lower cost, my model is, is better for this particular sovereign requirement. Like that's a very logical sort of outcome we would end up with in this, uh, in this landscape.

Speaker B: Yeah, I agree. And I think it's good, like it's healthy for a business because I, uh, think a few years back we were thinking that it's going to be kind of, uh, winner takes all, you know, where you have like one, uh, model that can beat them all. But um, now we think, yeah, I

Speaker A: think the past, the past M2 months with like the token costs growing, the, the, the sort of, you know, benchmark, uh, improvements we've seen from open models, like, and then, and then the fable thing as there are this extra kind of element that we didn't really factor in here. I think these are actually all very good things for the, for the overall AI space.

Speaker B: And I'm actually curious because, uh, you made like, uh, I mean you're at the forefront of AI and I've seen you talk about it from the start and you also made the transition like very quickly from SaaS to AI and to agent at Box. Like how has your vision like uh, evolved and how do you go from a pure SaaS business model to something that's leveraging AI and hence as also like different cost structure for your business?

Speaker A: Yeah, so, uh, so for us is um, uh, you know, we, we have um. Uh, our, our whole strategy is agents, uh, need access to critical enterprise data to be able to do their work. They need access to your documents, your marketing assets, your research, your contracts, your financial records. And so we built a platform that helps agents get access to that data and, and do useful work with it. So, so that's kind of the overall value proposition. We do this in basically two ways. One is you can use agents that are directly built on Box, and those agents use the frontier intelligence that we've been talking about. Um, it could be from OpenAI, it could be from Anthropic, it could be from Gemini, it could be from an open model. And we will plug into all of the agentic ecosystems, uh, that you want to use. Uh, so they can plug into Claude, Cowork and Codex and ChatGPT, uh, and Mr. All's chat system. So we plug into any AI system you want to use and we have our own Agents that let you do a variety of kind of document use cases and um, unstructured data workflows. So that's our sort of um, uh strategy. And the really exciting thing is what it does is it lets you just unleash all of these new capabilities on your enterprise data that weren't possible before. Um, uh, you can give our agent access to all of your research or all of your marketing assets or all of your internal policies or your entire product roadmap. And then any user can just go and either ask a question of that data or generate new content or generate a sales presentation or uh, go and process contracts and see where the risk is. So anything you want to be able to do with all that unstructured data, our agentic platform uh, effectively is uh, is powering for you.

Speaker B: And do you feel like the, the way people are going to work inside companies are going to evolve towards. Because I think like where you know like Box is super well structured is you have like all your documents into one place and you can like manage them but now you're kind of building the company's brain because you have like all these files that are like uh, you know like um, basically like accessible and queryable in real time with MCPS, etc. Etc. So do you feel like companies are going to move from a ah, place where you can really like access and structure documents? So for example, if you look at uh, at notion essentially a lot of people are writing their sops on notion writing like uh, guidelines for marketing, et cetera, et cetera. But in the end is that really useful when you have agents you can query directly? I don't think so. And I see like uh, Box very well positioned toward uh, a new category of just like uh, yeah, just business information in general. But I'm curious to know how you perceive this.

Speaker A: Yeah, I think your, your kind of uh, angle there is, is correct. And you know we kind of think about it um, uh, a little bit like how you know, people worked. So, so, so basically uh, you know, people work with you know, some roles you work with contracts, other roles you work with marketing assets, other roles you work with product documentation, other roles you work with invoices. And we do all that work and agents need access to that same data and that same information that, that we work with. And so I can't really create a separate universe just for agents to operate in because I as an end user need access to the same data that they're working with. And so this idea that there's a sort of A, uh, separated agent brain that it has access to. And I kind of lightly contribute to that. That's sort of unlikely to work. What has to happen is the agent needs to be able to access the same data that I'm updating or that my colleagues updated my customers updating, which means that you need some kind of shared content system or knowledge system or file system that the agent and the people have access to. And so what is better than just literally the file system that we're already using, but built in such a way that both agents and people can use it? And so within box, we've basically built that platform. We primarily started out building it for people, then we actually added applications to it. So we've already kind of built this machine user orientation over a decade and a half, and now there's just one new type of user, which is an agent user. And so the whole idea is, what if you had a single file system that people, applications and agents all leverage? Um, and those agents could be from any external system, like a cloud coworker and Codex, et cetera, or it could be from, again, my own internal agents that I've designed within box. We're kind of indifferent to that approach. We obviously kind of monetize them differently. Uh, but that's the whole vision. And so this idea of your company now has this digital brain for agents and people to all work together is, uh, what we're building. And then there have been things that we've had to meaningfully change. So, for instance, a year ago, we could barely spell markdown. Uh, six months ago you could click a markdown file and look at it. And now we let you edit the markdown and let you save it and let you give it access to an agent. And, uh, the next thing we're doing is making sure we've got really good HTML compatibility because you want to let the agent export a bunch of stuff that you want to go visualize. Um, so you have to improve the system to make it easy to work with and share and collaborate with the agent. Um, we've had to launch, obviously, an MCP server. We've had to dramatically improve our cli. We're playing around with new kind of file system ideas that agents can access. So there's a lot of underlying technology that's needed to make this all work. Uh, but obviously incredibly exciting for us because of all these new use cases that we can go empower.

Speaker B: Yeah, No, I agree 100%. And when it comes to the narrative, you know, that, uh, SaaS are dead uh, what do you think people get usually wrong with that narrative?

Speaker A: Um, I think there's two things that they get wrong, uh, and then one thing that they get right, actually. So let me start with the right thing so we can move to the positive. The right thing is that there is some SaaS that should get kind of compressed because at some point if I ask an agent to, you know, do X task, if X task used to be the thing that that SaaS did holistically, then I'm obviously going to use that, that SaaS system less. So that's sort of inevitable. So, so let's just say some SaaS will fall into that category and we can kind of carve that out. Then there's two reasons why agents, uh, um, are. Both are neutral to SaaS and then good for sass. So the neutral part, as in, like, as uh, in it, it doesn't tell you really much bad or good is that agents need the same access to data, guardrails, workflows that people need. And so what, uh, is the best way that an agent is going to look at a CRM record that I, as an end user have access to? It's the same CRM system that I already use. Um, it's the same ERP system that I already use. It's the same, um, document management system that I already use. So the ability to have guard rails and controls and to be able to govern what that agent does, that all makes sense as the same system that you've already kind of implemented or built out or that had to perfect it for that end user. Okay, so that's, that's sort of why it's at least neutral. Now here's why. It's actually positive in some categories. In a lot of categories, uh, these systems are actually kind of underutilized relative to their potential. Uh, and when you imagine a world where maybe there's 100 times more agents than people and they're all roaming around these systems and they're executing actions and pulling up information and generating new data, then actually these systems become even more important because they grow in the number of, uh, and range of use cases that they can go and execute. So for instance, within box, um, we have customers that will upload thousands, tens of thousands, hundreds of thousands of contracts, just to take one example. And those contracts are only valuable when like an actual person goes in and they load the contract and they look at it and they do control F and they look for the clause. That's historically how these systems have been used. Well, now all of a sudden an agent could go, first of all, it could go process all of that data and extract the most important intelligence. But you could also just say, hey, I need to look for like the riskiest clause of this kind of industry for these kinds of contracts that are due at this period. Now all of a sudden I can take all of this data in my enterprise and turn it into useful business value. That was not possible before. So then thus the system that organizes that data and connects agents to it successfully and securely, that grows in importance because now I'm using that data in all new ways. And you're seeing some early signs of what that looks like. So Snowflake, for instance, they just had a complete blockbuster quarter, uh, this past quarter because they're seeing kind of accelerating adoption. Uh, for us, we've meaningfully accelerated our growth rate because customers are upgrading into our enterprise plan that has this functionality. Um, and now they're using their data even more and in more use cases. So I think there's actually a lot of categories where as agents get deployed into that kind of work domain, the value of the underlying SaaS system will go up meaningfully as a result of all the agentic workloads that are happening.

Speaker B: Nice. Yeah, I agree with you. And do you feel like, uh, we talk also a lot about, uh, how A.I. is going to kind of like replace job and uh, we're talking a lot about graduate jobs and these kind of things that are disappearing. So what's kind of like your view on, uh, this?

Speaker A: Um, I think that there certainly can be some dislocation between the talent that's graduating, kind of what jobs they thought they were going into or what they just learned and then what the market has sort of available. And I do think we're at this sort of point where those two things aren't 100% connected. And so we have this temporary dislocation, uh, that for 20 years you, you got a CS degree and then you worked at Google or you worked at Microsoft or you worked at, at, you know, a Silicon Valley tech company. Like that was like a very straightforward path. And all of a sudden some of those companies are doing layoffs. Some of them are not as hiring as much. And so you have to kind of step back and you say, okay, where's my CS degree useful now? Well, you know, first of all, there's a lot of AI startups that are hiring like mad. Um, you know, I, uh, I know maybe 20, 30, 50 sort of startups and just generally tracking the industry. They're all hiring engineers because it turns out that you still need an engineer to manage the agent that's writing code. Uh, and that requires a high degree of technical skill to go do that. And so one, maybe you're not going to join Meta, but maybe you're going to Join cursor or OpenAI or an up and coming startup as an example. The next thing is maybe you look and you say, okay, maybe I'm not going to join a software company because what I'm going to do is I'm going to deploy my CS skills across all of the industries that now actually have more demand for software than they did before. Life sciences, industrial companies, manufacturing companies, they're all going to be hiring software engineers because AI coding has now made it so that those companies can basically light up way more projects than ever before. And so I think you see this slight dislocation, but that's different from the jobs not being there or, or not being available. If you kind of again go and tilt your skill set to that, that's, that's uh, cs. And I pick on CS because it's kind of the quintessential role that we think AI has already eliminated when we're actually seeing the opposite, where there's actually vastly more use cases for engineering. And so then you kind of go down the list and you say, okay, well what about, you know, salespeople? Won't I just, you know, go and sell the software? And guess what? You know, basically every single company that is building AI systems right now are hiring AI, uh, you know, salespeople like Matt. We have, we, we can, we can barely fill the amount of sales headcount that we have open because we're still, we're actually still in a talent constrained environment for uh, that type of skill set as an example. And so I just think that, that yes, there can be some dislocation for a moment, but when I look at sales, when I look at sales, when I look at marketing, when I look at engineering, when I look at more technical roles like, you know, biomedical and life sciences and industrial roles and physics physicists and a bunch of, you know, kind of critical industries, I don't think that work is going away. And uh, and I think there's gonna be plenty of opportunity to, to kind of take those jobs. And you know, obviously the work will look different because of agents, uh, but I think the opportunities will be there.

Speaker B: 100% agree. I know you're a busy man and we're almost out of time, so where can people like, uh, follow you and follow your updates.

Speaker A: Uh, well, you know, obviously I'm on, uh, I'm on X talking about all this stuff, but, uh, definitely, you know, pay attention to our announcements and the, uh, products that we're putting out there, and I appreciate the time.

Speaker B: Awesome. Thanks a lot, Darren. Have a great day. Thank you.