Aaron Levie, Box CEO: Advice for CIOs on AI Agents

There was a brief moment where you could rely on the subsidization from venture capitalists and then get tokens for your your, uh, your coding agents. It was a beautiful moment in history. It lasted about a year, and we all had a great time.

Every company wants AI agents, but few CIOs can run them safely at scale. Aaron Levie is co-founder and CEO of Box, which counts 68% of the Fortune 500 as customers.

We are still in the very early stages of what what agentic work looks like in the enterprise and what the rollout looks like. We have an interesting dynamic, which is sort of a tale of two cities. We have AI, uh, kind of agentic coding, which has clearly taken off, um, and it's within engineering teams, and everybody's kinda figured out what the the new practices are around the future of engineering, um, and we can kind of get into a a little bit of our own, what the differences are between the engineering side and the rest of work, um, but engineering, obviously, kind of complete vertical takeoff of AI agents.

And then you kind of get into the real messy environments of knowledge work, where things are just quite a bit different. They're it's a lot harder to deploy agents at scale. The agents don't have always access to the right data. The users are less technical, so they don't know how to sort of always steer them properly. Um, you have to do a lot of work to kind of verify the the work that the agent is doing in a way that in software, you can kind of just, you know, test the software and and see if it worked.

But you can't do that in a lot of areas of knowledge work. So I think we're in this really interesting phase where where we've seen what the promise of agents looks like, uh, in coding in in particular, we've seen what the promise of of kind of chat bots look like in knowledge work. And now the question is, what's the the full promise of agents across knowledge work? And this is the I think this will be the defining topic, certainly over the next few years within, uh, within the enterprise.

Let's talk about the promise of agents. Uh, just very briefly, first, talk about the promise of agents for programming. And then let's shift into what is that promise for knowledge work? And what are the gaps? How do we get there, and what's preventing it? What are the obstacles?

yeah, so the the promise of coding, uh, is honestly just an incredible gift, uh, which is, you know, we've all done large engineering projects, and we know how much time, you know, maybe sixty, seventy, 80% of of your engineering time is going into the work that doesn't really actually make the the project differentiated per se. It's necessary work, so it's differentiated in the sense that if you didn't do it, you wouldn't be successful. Um, but it's the the very labor-intensive, time-intensive things like, Okay, you know, how do we upgrade the version of this library to the latest version?

How do we, um, you know, make sure that that we're we're doing all of the edge case testing of our software? How do we go in and build all of the end user features that that are necessary for for this functionality to to be delivered? That's where most of the time of engineering goes, and, um, and that, you know, could mean that you wanna take on a big project. And it, uh, you know, just that from a standing start could take six months or a year or two years before you see the value, um, of that product delivered to your customers, whether it's an internal solution or a customer-facing solution.

So the promise of agentic coding is, well, what if we could have agents go and do lots and lots of that that blocking and tackling work that that's necessary? And then our job is making sure that we're we're giving the agents the right plan to work with. We're steering agents in the right direction. We're reviewing the work of those agents. We're we're deploying, you know, testing and security solutions, uh, for for whatever they've worked on, uh, and could we shrink that one-year project into two months?

Um, and that was a kind of a fantastical concept that that honestly would have, you know, people would not have believed you three or four years ago if you had said that. Cause all we had was like, type-ahead functionality in, um, in our in our code editor. Now that promise is reality, like we're seeing this every single day internally at Box, we're seeing it from our customers, which is we might be able to literally do, you know, three, five, ten times more work, uh, than we were able to do before.

And the corollary is you might be able to do it three or five or ten times faster. Um, so this is what we're seeing on the engineering side. Uh, at Box, it means that we can just build a much more significant product roadmap for our customers. It means that they can take on way bigger problems than they could have before, which is, uh, which is amazing. So that that's the promise.

This episode is brought to you by Gartner IT Symposium Expo Ready to scale agentic AI from pilot to production. Join top CIOs and IT execs this October nineteenth to twenty second in Orlando, Florida. Over 300 Gartner analyst-led sessions will cover top priority shaping IT from AI value, governance and cybersecurity to cost optimization, IT operating models, and beyond. Get practical insights and connect with peers tackling the same challenges you are. Secure your spot today at gartner dot com slash us slash symposium.

What about knowledge work? Why you you you describe that as being messy, complicated?

yeah, yeah, so so that's the promise, and actually what we're starting to see the early signs of in in the engineering side. So now everybody kinda sits around and they say, well, we want that exact same kind of output and outcome in knowledge work. And, you know, you have a you do have a couple differences that that have to be figured out. Um, so the first is that, uh, I'll just go through maybe some of the delta.

In engineering, you've got a lot of great properties for for agentic work. The work is mostly in a text-based medium, so, you know, you're mostly just writing code. Um, AI agents have been trained on this data, you know, almost, uh, almost, you know, from a plurality, uh, of of the data that they get trained on. It's it's lots and lots of of of code examples, uh, on the internet. Um, they, uh, your your users are obviously much more technical because they're engineers, and so they can they can, you know, both implement the technology themselves, but they can keep up with all of the updates that are happening in AI.

So it's much easier when a new model comes out that they can go and implement that latest model, um, or when some new kind of alpha emerges in the industry, where, you know, if you configure MCPs or CLIs in a certain way, they can adopt that much faster. Um, you have another property, which is that the the work is is more or less verifiable in that in that if I build a bunch of code, uh, and I can go and do a regression test on it, I can see, you know, did I break anything or does it still work?

Um, now compare that to knowledge work, um, you know, you have a, you know, generally less technical audience just by definition. The work is sort of less verifiable by by definition. So a contract, you can't, like, compute whether the contract is sort of correct or not. It has to actually sort of, you know, be be dealt with. It has to sort of experience reality. It has to experience the red lines from the other party. It has to experience reality of of somebody, you know, took took your, um, uh, took that case to court.

So you don't have you have to then review the work a lot more to make sure that that does it actually sort of do the thing that you that you wanted it to do. You have to review the the final financial analysis to make sure did it pull all the right data in the right way, you know, from your set of spreadsheets, um, and the agents don't have access to data in exactly the same way that that they have access to the data inside of engineering.

So, in engineering, you know, from an access control standpoint, in general, uh, if you're an engineer working on a project, you already have access to the entire code base relevant to that project. And so by definition, the eng the agent that you're deploying also has access to all of that data. Well, in the enterprise, we're constantly asking for permission to other systems and other resources and other data environments. And so an agent is only as good as the data that it has access to.

But in the enterprise, we have lots of systems that are either not well maintained or the agent can't get access to the right data, or maybe even worse, you have too much access to information. Um, and we have sort of, s you know, a security through obscurity, uh, in your organizations, and now all of a sudden the the agent is leaking data, uh, to the wrong people. So think about just that that sort of T chart that I just went through, um, and think about how many differences there are between agentic coding and agentic knowledge work.

So the project ahead for all of us, uh, you know, kind of across the economy, whether you're a tech company or you're, you know, a a user of technology, is how do we take those same gains that are happening in coding and how do you bring them to the rest of the world and the rest of our organizations? And that is just going to be a a very big project, uh, that will take years and years of of sort of diffusion into organizations.

Now I'm extremely excited about it because it means there's all new roles that are emerging to go do this. These are the the upside when you're on the other end is is tremendous because now you can actually accelerate your your organizational productivity. But but make no mistake, this is a this is a very real effort that that companies have to go through right now.

Folks, you can ask your questions. When else will you have the chance to ask Aaron Levie, the CEO of Box, pretty much whatever you want? So take advantage of this opportunity and ask your questions. Aaron, you recently said that CEOs are uniquely prone to AI psychosis because, and I quote, they're sufficiently distant from the last mile of work that still has to happen to generate most value with AI. Can you unpack that for us?

I sort of have my own experience of AI, and then I've seen it from from either peers or or other people in the industry. And so I I kind of I believe I have a good a a good sort of pulse on this, which is, um, I I use AI, I think, as much as as any CEO out there. Um, uh, I'm I'm using it for, you know, prototyping of new products, uh, developing new ideas, doing lots of market research, um, automating, you know, customer intelligence.

And so I'm using all AI all day long. And, um, and at the very start of of that journey, you do these things like you you prototype a piece of software or you do some market research, and you are like, Oh my god, this thing is going to just automate everything. And and you know what, what are the implications of this and what does this mean? Um, and you almost have this sort of existential dread, you know, for for a few moments there, and and you know, s for some people, it it lasts longer.

And I think I I maybe got through it, maybe, you know, a couple weeks. Um, and and by the way, it kind of ebbs and flows because a new model comes out, and fable fable emerges, and you're like, Oh my god, I'm experiencing it again. I'm about to I'm about to get into the AI psychosis mode. And so there's sort of a a juncture, which is which is, you know, one you can end up on one end of the juncture, which is, oh my gosh, like, we have to retool everything and and what are the people going to do?

And and, you know, it gets very existential. And the other end is is you start to kind of see, like, oh, no, like, actually, there's still lots and lots of of bugs that the agent, you know, wrote. There's still lots of security issues, uh, that were were generated or or even maybe discovered, uh, by the agent that now have to go get fixed. There's all the ongoing maintenance of the system that we just deployed. There's new new features that we obviously want to go in and work on.

That's the coding version. But in other areas of knowledge work, you find, you know, sort of things like, oh, the agent pulled from the wrong piece of data, which meant that if I hadn't reviewed that report, uh, I would have come to exactly the wrong conclusion. And so I had to actually steer the agent a bit more, and it took a lot of work to kind of prompt it in the right way. So on the other sort of, you know, fork in the road, you end up, you know, kind of maybe post psychosis, which is which is this period where you realize that, oh, actually, like, there's still a lot of work that has to get done to make these

agents be effective. And, uh, and so this is sort sort of journey that you go on, which is this inc- you know, instant sort of existential dread or high, which is, oh my gosh, what is this going to mean? To then a bit more of kind of a pragmatism that that sort of sets in for for at least that's kind of where I landed, and I think, you know, I have a number of of friends that also landed here, which is like, oh, okay, actually, this is actually a technology that is going to boost our productivity.

It's going to let us do far more than what we were able to do before. Um, but it is not as, uh, sort of, you know, you know, doomsy as as I think some of the commentary, you know, comes out. Because actually, it's just gonna it it's gonna take a lot of time to fully be able to deploy these agents, to be able to get the work done that we need. And so that last mile thing, or the kind of closeness to the problem is is really the issue.

Is like a CEO, by definition, is the furthest away from the real work that's happening in the company. Like, you couldn't get further from, um, you know, in any other role other than maybe the board of directors. And so, uh, and so for us, it's very easy to be like, oh, well, like, I could just automate that engineer, uh, or I can go automate that marketing campaign. But then when you're closer to the problem, you realize, like, no, actually, like, like, you probably just can't have a uh uh, an agent go and do all of that without any su- human supervision because it's gonna do the wrong thing or the the taste of what the agent delivers is gonna be

off or it's gonna introduce a bug. So that keeps the humans in the loop, you know, really, as far as I a a as far as I can tell, for the foreseeable future, that that's gonna maintain the case. And so, uh, I think the thing I encourage CEOs is use the technology actually so much that you get to the other end of that psychosis. And you can actually see, uh, in a much more practical and pragmatic way, all the places where humans are still necessary to really get the ultimate gains from this technology.

And don't just stop when you just prompt an an agent and you just ask it to to generate some research report and then you get blown away and then you you sort of think, well, that's that's just the only thing that's going to happen in the future.

This episode is brought to you by Gartner IT Symposium Expo Ready to scale agentic AI from pilot to production. Join top CIOs and IT execs this October nineteenth to twenty second in Orlando, Florida. Over 300 Gartner analyst-led sessions will cover top priority shaping IT from AI value, governance and cybersecurity to cost optimization, IT operating models, and beyond. Get practical insights and connect with peers tackling the same challenges you are. Secure your spot today at gartner dot com slash us slash symposium.

I think that there is a huge challenge in terms of calibrating what agents can do, what they can do safely, where you can leave them alone. And this is where people have a hard time. And on this topic, we have what we have a bunch of questions that are coming in. And let's start with Gus Bechtash, who is the brave one who asked the first question on Twitter. And Gus says this, he says, uh, some projects declare amazing victories because their objectives were so modest.

How do you determine the right level of ambition to avoid irrelevance by being too small or too large in AI and automation in general?

my instinct is always biased toward, uh, being more ambitious with what you can do. Uh, that will more often lead you to seeing, um, uh, first of all, better productivity gains, but then also, you know, see the edge of this technology much faster. And and I do think that I think anytime you hear stories about people not getting real ROI from agents, it's I do think it it often approximates, um, you know, not not pushing them hard enough.

And thus, you know, doing very kind of more menial type of of work as opposed to let's like, let's actually push the limits of this technology and and um, a and really kind of, you know, get the gains out of it. Um, so I don't have it's it's hard to have like a a a full general piece of feedback other than I would just always bias toward being more ambitious, uh, pushing the limits much, much further.

And then and then, importantly, uh, if something fails with AI, uh, with whatever that ambitious idea was, maybe a big software project, maybe some, you know, go-to-market, uh, automation activity. You almost have to try it again six months later, um, like almost every single time, no matter what it is. Because the model progress that's happening right now is is still actually, uh, an in an incredible rate of progress that we're seeing. So you often almost have to say, you know what?

I have to forget about the fact that six months ago, this this didn't work in this particular part of our code base or in this particular marketing campaign. You know, if you relied on your understanding of image generation as a marketer from six months ago, that's already been blown up. Like, the latest image gen models, uh, are are perfect at being able to do text. They have photorealistic, you know, capabilities in a way that wasn't possible six or twelve months ago.

And so if you were in, you know, marketing campaigns, you'd want to kind of reset your understanding. If you're in software projects, something like Fable or GPT 5.5 actually is another step function improvement on what these agents were capable of. We we had a project just yesterday where we tested Fable versus Opus 4.8. And you'd you'd think, like, okay, these are should be on the same exact trajectory of of progress. And the thing that Opus 4.8 couldn't do, Fable just finally actually did.

And so if you don't actually constantly recheck what is now possible with AI, you will fall into the trap of of sort of, you know, kind of believing that that maybe the productivity gains aren't there, or you're not able to accomplish as much as you want. So I I think have very ambitious expectations, constantly be trying the technology, um, and and, you know, keep pushing the limits.

It is incredible how fast things are changing and the capa- how how rapidly the capabilities are improving. Okay, let's jump to some more questions. This is from Chris Faulkner on LinkedIn. What are your thoughts on the forward deployed engineer F D F D E concept?

I actually think there's going to be a role for effectively an internal FDE. Um, and this is some kind of of IT business, AI automation engineer type role. Uh, I think often it's going to live within the technology or IT organization, but be embedded, uh, within the actual line of business that it's it's trying to to bring automation to. So there's the internal version, and then there's the external version coming from either the vendor or the systems integrator or maybe a new consulting firm, which is effectively, you know, a as we if we go back to what's the difference between coding agents and the rest of knowledge work, it's it's a lot of it does come down to how technical

the user is and their ability to navigate around all of the challenges that agents run into. So one of the big challenges is, you know, do agents have access to the right data to be able to work with? Uh, do they have access to the right internal systems to be able to pull context, uh, into the context window to be effective? Well, an engineer kind of knows all their way around those different systems. They know how to get access, uh, to to to resources, um, they know how to set up MCP servers, they know how to set up CLIs, they can deal with kind of, you know, compute sandboxes that the agent might need to be able to run into.

This is just not necessarily how most of knowledge workers were trained. These are not the things that we've ever had to think about. You know, we we're we're kind of used to, like, we go into PowerPoint, we go into Word, we go into GDocs, we go into Slack. We we we generate, you know, a a bunch of stuff. We know how to create files and share them. We know how to to do, you know, kind of lightweight manipulation of these tools.

But but actually, for an agent to be really effective, they need access to all the right data. The data's got to be in the right format. You probably need skills files and, uh, and ways of having agent sort of, you know, kind of read me files in in the process. This is sort of this agents dot md phenomenon that it's emerged. Uh, you have to have the permissions, uh, to actually, you know, turn on the MCP servers.

All of that is technology work. There's almost no way for that to not be done by somebody at least either formally trained or has sort of figured out how to how to be highly technical. So the job of FDEs is is basically go help companies, uh, sort of, but first of all, understand today's workflow and then figure out how to how to bridge the capability of the technology that is emerging and and constantly changing with the workflows of that organization.

And sometimes you can get lucky and you can kind of embed agents in an existing business process and and it it sort of just miraculously works. But oftentimes, you actually have to reengineer the process. You have to you have to migrate the data sources into a modern system. You have to sort of change the workflow so agents can be more effective in that workflow. All of that is effectively the work of the Ford deployed engineer. And that Ford deployed engineer could come from a vendor like a Box or a Salesforce, or, um, or a Palantir, or it could come from a systems integrator or maybe a new consulting firm.

So I think there's going to be all these new roles that emerge, which are basically the technical talent that has enough business acumen to to get into the business processes of that organization, but successfully implement the tools of of AI, uh, to to to actually bring agents to to bear on those workflows.

So one of the very interesting aspects of all of this to me is you've got the technology on the one side. And then, as you were just describing, you have the organizational ramifications, the implications for the processes, for the talent, for the skills, for the mindsets, for the culture. But Robbie Hassan on LinkedIn says, looking ahead three to five years, but I'm going to say looking ahead for a year or two. Do you think the biggest competitive advantage will come from having the best AI models or still owning the best proprietary data workflows and distribution?

And I'll ask you to also incorporate into this the extent to which the models are becoming commoditized because you just yourself just described, you know, the using different models, which means you're swapping.

I think we're still at a point where I would bet on either, uh, a technology company or an end user of the technology that has simultaneously, you know, the frontier models that that they're leveraging and the best, uh, ability to get those models the right context to to be able to work with. And so, you know, kind of s s sometimes you kind of want to do this thought experiment, which is if you were starting your company from scratch, what would it look like to be able to get the full gains of of AI?

And I think it's kind of, you know, pretty intuitive that if you could start your company from scratch, you'd basically design your business processes in such a way where the agent has sort of innate ability to get access to the context it needs to to help you automate work. And so, like, like, if we we if we if we started a one-person company, um, and and you know, you'd probably start with, like, a file system that had all of the context of of your business.

It had all of your goals, it had all of your your your marketing strategy, it had all of your, you know, customer, um, uh, sort of language that you used. And then you'd build your company up from that foundation, which is okay. Agents always need access to the right data to to, you know, answer the right question or be able to automate the workflow. So if you were to build your company from scratch, you would do both the frontier intelligence and you'd have proprietary information and insight that you're giving that agent, and you'd kind of develop a flywheel.

And so then as you hired more people in, you would try and sort of maintain that ability for agents to keep having access to the right data. You'd really value sort of sources of truth and authoritative sort of knowledge banks. Um, you know, one of the problems that that traditional enterprises have is our sources of truth are everywhere, and many of them are not well maintained. But if you were starting from scratch, you would just always be like, you know what?

We always write down our final decisions in a way that agents can ac- get access to. So that way, the agent doesn't sort of ra- you know, land on the wrong resource, or it gets the wrong insight from from some data we give it. So that that's what you do if you could start from scratch. To so to the question, you know, what, what should you do right now in any kind of meaningful or or large enterprise?

I think you still you basically are trying to do whatever you can to get to the point where where a company from scratch, you know, it it sort of is built up in this kind of way. So I would be betting on frontier intelligence right now, with the combination of access to the right context for those agents to be to be effective and making sure that your your organization is built in such a way where you can get those agents the right data in those workflows.

Now, to your second question that you added on top of that, this is sort of a different, you know? And then then the question is, architecturally, does that mean that you sort of bet entirely on on one frontier model? Probably not. You probably want to to sort of design some some kind of neutrality where right now, I think you should still try and exploit the gains from frontier intelligence as much as possible. But in one or two or three years from now, I think you're going to see this sort of stratification between the cost of frontier intelligence and the cost of sort of the second best frontier intelligence.

But that can do the job extremely well and in a high quality way for a bunch of sort of long-tail tasks. And I think you're gonna end up in a world in, you know, one, two, three years from now where model routing becomes very important, where maybe something like a fable-esque model gets the planning part of the work and the review part of the work. But the in-between sort of massive token usage comes from maybe something that is a a more cost-effective model for that type of work.

So then thus, you want to invest in the right kind of architecture that can deliver on that type of outcome. Uh, that's that's probably where we're heading from a from a kind of tokenomics standpoint. Uh, but but I think, you know, making sure that you can de- design an IT architecture and a data architecture that lets you deliver that is is super important.

Such a great, important point about the designing of the right type of flexible architecture that lets you accommodate changes in models, uh, both in term both in terms of the capabilities and the costs.

I thought this was true in the kind of cloud and SaaS era, but now it's like an order of magnitude more true. We we used to have this kind of thought process, which is if you show me your your IT stack, I'll I'll show you your culture, um, which is which is, you know, if I I can tell you from the tools you use, the the kind of culture you have and and how fast-paced your your company can be, um, how innovative it is.

Because you can just see, like, it does the average employee have access to the data that they need to do their work? Um, can they communicate, you know, instantaneously with their colleagues, with with very, you know, few, um, you know, few few barriers? So that was like kind of a ten, fifteen year ago version of this when you when you sort of saw the rise of Zoom and Slack and these kinds of of solutions. I think there's a new version today, which is if you show me your IT stack, I'll show you what you're going to be able to get from agents.

Um, I and I can just you can just see it instantly. What what kind of agentic productivity and outcomes you will get based on where is your data today? How much is in legacy systems? How much are in systems where the agent has has access to these, uh, tools? What what is the frontier intelligence you're using? And so I I I think, you know, oftentimes we're having an AI conversation, but it's sort of masking an IT architecture conversation or a data architecture conversation.

Uh, and that's really a lot of the conversation we should be having, which is like, do you have the right data platforms? Do you have the right IT architecture to let you get the real productivity gains from AI right now?

You mentioned Aaron, the term tokenomics, and we have a question from Chris Peterson on Twitter X, who says, How is Box dealing with model providers and others increasingly moving from subscriptions to poor toc- uh, pure tokenomic chargeback? In other in other words, how are you? How does Box deal with these increasing costs?

We definitely preferred, uh, the world where it was being subsidized, um, so tho- those days are over. Um, you know, it was, uh, it was, it was fun while we got it, guys. And, uh, and it was, um, uh, you know, there was a brief moment where you could rely on the subsidization from venture capitalists, uh, and and then get, you know, kind of tokens for your your, uh, your coding agents.

That was a beautiful thing.

It was a beautiful moment in history. It lasted about a year, um, and we all had a great time. Uh, so, you know, there was definitely actually a period there where where you probably could have, like, really exploited it and, like, had, like, ten years of software development, you know, paid for by, uh, by by, uh, LPS and and VCs. Unfortunately, those days appear to be coming to an end. Um, uh, but now we're, you know, now we're in a much healthier environment, which is you're you're more or less gonna be paying for for the real underlying costs that that it takes to deliver this.

And then I think it's just capitalism, you know, microeconomics one zero one, which is which is, you know, wh- where? where are you getting the right ROI from these agents? And and you should deploy, uh, you should deploy agents at the work where, um, where it's most effective, where you are getting that ROI. Um, you know, we're we're fortunately in a little bit of a sweet spot where, um, because we're an enterprise software company, because what we build, we we have a pretty good pulse on our customer base.

So we kind of know how to deploy, um, you know, agents in a way that makes us more productive and lets us ship more software that we think is gonna be valuable. Um, uh, so we are we're kind of smack in the middle of like, if you have like a token maxing company, you know, we're not that. You know, we don't have a leaderboard internally, we don't try and incentivize the most number of tokens. And then on the other end of the spectrum, maybe you'd have a company which is like everything's locked down.

It's like really hard to get access to frontier agents. We don't want to be that either. Um, we try and rely on, you know, more, um, uh, sort of mature, sophisticated approaches of like, as long as we have a really good product roadmap, uh, and we have really good sort of product managers and engineers, kind of, uh, you know, working on on designing that, then actually, I'd rather move that product roadmap forward as fast as humanly possible.

Which means that that if if the token spend goes up exponentially, uh, that that should actually be correlated with a good thing, which is we can deliver more software to our customers faster. And then it's more of a CFO exercise, which is, how do you plan for that? How do we budget for that? What are the trade-offs we're making in the organization? Um, I do think the the corollary to this is, um, is we're well past the point where the IT organization can hold the entire budget, uh, of AI.

Um, and, uh, and that that's actually a healthy thing, and it's a good thing for this as a technology, which is, you know, IT budgets kind of run at depending on the industry, maybe three, five, you know, you know, two to two to 5% of of sort of total revenue, um, depending on the kind of industry that you're in. And, um, and that's obviously an a sort of an artificial cap on what AI AI's potential could be.

You know, maybe you want AI alone to be 5% of your your total revenue, so that would be a doubling of the IT budget. Well, to do that, then you ultimately need the line of business to own the budget, um, and own the the sort of deployment of where do they want, you know, AI being used in the organization. So I think we're in this renaissance, actually, of the IT organization, which is, uh, your job now is to bring kind of, you know, some form of intelligence to the entire business, um, and be the the sort of experts in the technology and what what you're going to be able to deliver with that technology.

You're going to partner with the business on where should that get deployed, um, you know, and then that's that's sort of this interesting budget slash use case meets technology and capability kind of pairing that we're going to have for for quite some time.

We have three questions from Maya Cunningham, Abdullah Alganimi, and Aga Salman, and they're all asking, How should knowledge workers prepare for an agentic future? What are the AI automation roles? How should they get started in automation? Which tools?

I think this is an incredible time if you're if you're super curious, if you're excited by technology, if you're one of these people that geeks out on, you know, the latest version of of the iOS release, and now it's the latest version of the the Cloud CoWork release, um, or Codex release. So so I would I would go very deep in the technology. I would be experimenting with it constantly to kind of push the limits of what these tools can do.

The awesome thing is that, you know, uh, it it's one of the first moments in history where where a almost everybody, and you know, there's matrix, but but almost everybody has the same amount of access to information, um, uh, at the same moment. So, you know, twenty years ago, there was this huge advantage where if you were in Silicon Valley, you you, you know, you would see the iPhone first, cause you could go to the developer conference and and look at it.

And, um, you saw what was happening in SaaS, cause most of the companies were located here. Today, uh, in 2026, there's like, you know, maybe I have an hour, you know, heads up of information from from anybody else. Cause like there's some chat thread that that's going on with Silicon Valley founders or something. But other than that, we all have access to the same technology effectively at the same moment. Which means that that if you're paying attention to the right resources, you know, this this podcast being one of them, um, you know, there's other great AI podcasts.

Like, you can be as informed as as the best expert in the world right now. You get the same news feed as Andre Karpathy, you get the same news feed as as Greg Brockman, uh, or Dario Amodei and and Sam Altman. So if you're getting the same news feed, then that means that there's very little that they have access to other than maybe what's in their research labs that you can't take advantage of. So I think we have this in cool, cool moment in history where the information sources are keeping up with the pace of the innovation.

Doesn't mean our organizations are, but but it means that our our our ability to tap into what's happening. So I would just be playing with this technology constantly, I'd be I'd be pushing the limits, I'd be breaking things, um, and then I would I would try and figure out, like, what is this? And what are the implications of this technology to either an organization I already work with or an organization I want to work in? And and I think that those kinds of of people and personalities will be the ones that that, you know, for now, especially, get ahead, you know, uh, uh, farther.

It's so true, you know, I am not a developer, I'm a typical knowledge worker, but I force myself to use Claude Code to use it, to learn it, and the things that I can do with coding, like, for example, on our website that I've wanted to do for years. And it's it's not only the labor, but the agent has access to specialized knowledge that one person cannot possibly have. And so, instead of having a team of specialists, my agent is doing various things.

I mean, I go, it's it's unbelievable.

It's a key insight that that some people miss a lot because you're always so used to thinking about this as kind of like, like, you know, we we sort of anthropomorphize this technology, you know, probably sometimes too much, but but it's not. It's not just an engineer or just a marketer, it's got this ability to have the all of the skills in one. So so I'll often go to AI, uh, for product prototyping, and and I I might give it a prompt, um, uh, that that is sort of very specific, but sometimes I'll actually add like, and feel free to add anything else you come up with, uh, into the prompt.

And so that'll be an explicit, you know, sentence or two in the prompt. And it will it will come up with an idea that far exceeds what I ever would have been able to prompt it to do because it can take in all of these extra domains that it knows about. And so if you're doing product prototyping, you know, try not to just limit to your own imagination of what you're trying to accomplish, you know, give it that extra nudge that says, hey, if you also have better ideas than what I'm telling you to go do, come up with it and show me what you're thinking about.

And the cool thing about AI, I mean, you know, it's getting more expensive, there's no question, but the cool thing about AI is you could just like, have five tabs running and have it go f- do five different versions. So I'm often prototyping in parallel just to see are there things that that I'm missing and can this expand the kind of, you know, use case or capability that I was trying to to kinda come up with?

Now would be an excellent time for you to subscribe to the CXO Talk newsletter so we can notify you about upcoming shows and you can really be part of our community. Go to cxo talk dot com and sign up for our newsletter and do that now. Okay, we have a very interesting question from Swami Vaidyanathan, who says, How do you see business models change as you bring agents? Overall, where do you see pricing pressure due to simplification and where governance costs nullify the benefits, if there are any?

This is a a big question in general across the software ecosystem, which is, you know, what's the new value proposition of software in this world of agents? And we've been pretty clear that, um, you know, I I think that that people can certainly have different takes on this. I I think that, you know, for the most part, you're not going to go and vibe code, uh, a CRM system or an ERP system. You know, w- w- you know, we don't think that you should be vibe coding like a core system of record for your documents and enterprise data.

Just cause just cause, you know, first of all, you're benefiting from the fact that there's also another 100,000 customers that need the same technology. So it's the job of that vendor to get really, really good at doing that one thing really well. Um, and, uh, and if anything, you know, with AI, we can actually do that for you even faster and and even better. And so so I think there's been some kind of confusion about where we're probably going to be applying the these these tokens and these capabilities.

So I think the value of kind of the core systems of record, you know, continue to to matter a ton. Um, but then the question is, what can you do with agents on top of those systems of record, and where can I get even more value from my technology? Uh, and this is what what I get really excited by because I think what's going to happen is we're just going to be using, um, our tools far more.

I'll give you one example. I use Salesforce more today than at any point in history, maybe like five times more. Because I have an MCP server, uh, I have their MCP server connected to Cloud CoWork, and I'm constantly banging on the MCP server, doing various customer intelligence, market intelligence that I would have never done by manually going into Salesforce and kind of pulling up records and and and looking at them. And so I actually think that there's been a little bit of a a kind of a misread on on the market opportunity, which is actually we're going to move to a world where there's going to be, you know, maybe a hundred times more agents than people using software.

So in a world of a hundred times more agents than people using software, what are we going to use these tools for? We're going to use them for so many more things in our organization. And we have an interesting lens in this because we get to see from our customers where they're just like processing data at a scale that never would have been possible before with humans. And so all of a sudden, like, like, the value of your unstructured data assets have have gone up.

The value of your your CRM systems have gone up. The value of what you can now do with your ERP system has gone up. So so that's, I think, the, you know, how the maybe the value shifts over time. Um, and then, you know, things like governance, all of these things become really, really important for for actually, then, you know, making sure you can effectively get those gains, uh, you know, from those technologies.

This is from Somak Sen Gupta, who says, Internally, within business functions, how are you reimagining workflows using AI? For example, you're looking at traditionally overlooked candidates like finance. And I'll mention we recently had as a guest on CXO Talk, the CFO of HPE, talking about how she's using agents in the finance function.

I think you should really, really treat this as a as a technology for abundance. And, um, and you almost have to kind of look at your business and and stop and ask yourself in every part of your business. If I had, and it's just good for a thought exercise, it's you know, you can't actually afford this, but but, like, if I had unlimited capacity in, you know, XYZ area, what would I what would I do differently in in that area of work?

Um, and, you know, unlimited, let's say, you know, capacity for for combing through information, for using judgment, for accessing data. If you could do that at an unlimited amount, what would change about that? About that workflow? And and, you know, I think you quickly can start to think through well, wow, like, there's actually a lot of parts of my business where I've been constrained by my ability to deploy human resources at certain problems. And if I get incrementally unconstrained because now agents can go do a lot of that work, what would I do differently?

So, you know, e- easy example in our kind of world, and as a B2B company, is like, if I could deploy agents to go and and comb through our customer base, we would have much better insights about the right time to to have that right message for our customer when they're dealing with with some issue that we can be helpful with. Um, if I could deploy unlimited compute, it would, you know, it would know everything that's going on in every single one of our customers, so we can be that much more helpful, um, in, uh, in in our relationship or our partnership.

You know, I think you you in your question, you asked about HR. Um, I posed this question to our recruiting team the other day. I said, you know, if you could just comb through all of, let's just say LinkedIn, and you could you you didn't, and and in instead of just sort of stopping at at, you know, the the LinkedIn profile. But you could then hop over to the internet and see, like, what were the GitHub projects that that person worked on?

Um, you know what, what's their thought leadership that they've that they've talked in the press about? Um, and you could build a sort of full profile of that of that individual. And at the right moment, when something's going on in that organization, that was the time to to either talk to them or recruit them or or have a conversation, you know, how would that change recruiting as an example? And so you can kind of go through, you know, your entire organization and and see where is your business?

Where could your business actually have completely different returns? Um, with if you had just unlimited capacity to go and and work on work on whatever whatever it is and be able to have unlimited information. And so, you know, to a CFO, it's, you know, there's a there's a huge wealth of examples, which is well, what if I could analyze, um, my business totally differently? What if I could see where maybe there's waste, you know, in the business from a operational, you know, expense standpoint?

I can go and take those dollars and reapply them to areas where we need to drive growth, or which parts of my my customer base are unprofitable that I can go and kind of tweak the business model for. Um, you know, where are there insights in in market opportunities that we should be doubling down in? All of these things, you're basically constrained today by the amount of people you have to work through spreadsheets and work through ERP systems and work through, you know, analytics data.

Well, now you can actually throw compute at that problem, and and all of a sudden you're no longer constrained by the number of people you have on the team. So so I think we're we're in this real renaissance again of of the kind of workflows that we can go in and, um, uh, and and execute, uh, and and. But that really does take kind of looking across the organization of where where would you have a lot more upside if you could bring, you know, u- unbounded resources to those areas.

This is from Santosh Vasanta Kumar, who says, What are some of the best practices followed at Box to measure value generated from AI adoption, for example, value maxing versus token maxing?

Definitely a big fan of of value maxing, and I'm glad that that that, you know, terminology has already taken over from token maxing pretty quickly. Um, I think, honestly, it's it's hard to be, um, too general, generalizable about this. I think it's probably no different than, you know, if you were to ask that exact question, um, twenty years ago, before before any form of AI or cloud technology, whatever. You just said, you know, how does somebody measure the ROI in the marketing team or in the sales team or in, um, in, uh, the finance team?

I think you'd have to use whatever tools we we have always thought about, which is which is, well, I have a certain amount of dollars. I can deploy those dollars against, you know, a variety of different sort of things in my business. I could I could do events, I could do, I could, I could, uh, I, I could deploy marketing campaigns, I could build products, I could hire people at a certain price point, I can, I can do infinite set of things.

So how have we ever measured ROI? It's it's basically amongst all of these choices and all of these opportunity costs, where is the most, uh, effective use of the incremental dollar going to go in? And the maybe the the bittersweet news here is is that it is no harder or easier to do this with AI than at any other point in history. It's always been this kind of squishy thing where there's some judgment and and you're you're kind of like, you know, sometimes you guess, and sometimes you guess wrong, and sometimes you guess and you guess right, and then you keep doubling down.

AI is no different. You know, we're you're going to be experimenting. You you need y- you know it, it the you need high judgment people. Maybe the one difference, I guess, is is certainly if you like, you know, with the wrong prompt and the wrong limits, you could probably, like, go and spend 50,000 and wake up the next day and, like, have that bill. So, like, you do have to be, like, pretty thoughtful about, okay, where are we going to go deploy these tokens?

Um, but short of just making sure you don't make, you know, catastrophic mistakes like that. I think the problem looks pretty similar to to the history of of business, which is you need people close to the business with budgets that have high judgment. They understand the technology, they understand what it's capable of, and then you need a rigorous process to constantly be reviewing where are you getting the ROI, you know, from from these deployments. This is not a sort of a one-stop, one-shot kind of environment.

This is an ongoing budget management process.

I have to highlight this comment of Aaron's because if you're a CIO, it's not just a matter of the technology. Success is not just a matter of the technology, but of developing. He used this term judgment and the business acumen. And this is not a new story. This has been going on, you know, forever. It's never it's never been more true than today. It's been true in the past, equally true today. Okay, this is from Zoe, uh, Feral Rodin, who says, Given the trap one falls into if you don't keep up with AI and the many iterations, what do you see as the most valuable skill?

Knowledge workers, the s most valuable skill knowledge workers need to have in today's world other than adaptabil- a, uh, adaptability.

I do think it's really important to to keep up with the tech technology. I don't know of a of a replacement to that, unfortunately. I know that was in as the qualifier to the question, but I I think you just have to find a way to stay as as current as possible. And it behooves companies for educating their employees as well on this.

And let's go to the next one. Greg Walters says, Uh, it seems that the best AI implementations rise from employee up versus the C-suite down. Your view on that?

I think that's like ninety accurate. Um, I think that there's, uh, y- y- you know, the the person that actually owns the delivery of a of a particular project is in the best position to to know, you know, what's the what's the rate of productivity they can get with AI, assuming that that they they have a religion on this. The only exception is when you just have some things that are either so expensive or transformational that you need the kind of senior leadership to get behind or or kind of to identify.

But but I would I would generally bias toward that point.

And this is from LinkedIn from Yah, who says, Aaron, building on your point about playing with the technology. If everyone has access to similar AI tools and information, what separates a strong early career candidate from the rest? What should students practice or build to demonstrate real AI fluency?

Again, this combination of both technical skill and and business acumen, um, is is sort of still the best approach, which is which is if you're interested in marketing, get both really, really good at marketing and the and the core principles of marketing. But also understand how AI accelerates building a marketing campaign or doing market research. And so being able to live at that intersection is, I think, still the the most sort of potent way that you're going to be able to go and deploy this technology.

And so I would I would tell that to, you know, if you're in, if you're doing, you know, clinical, uh, drug drug trials and you're in life sciences, you know, deeply understand that. But then also understand how something like Claude Code or Codex helps accelerate that kind of work.

And Simone Jo Moore says, Do you see governance changes now as an AI add-on or embedded? And she'd love you to use the word squishy.

Okay, squishy, uh, squishy, for sure. Uh, I I think it's gotta be, um, I I think it's mostly embedded, but there's some some areas where it's gotta be, you know, kind of an adjunct.

Ken Walker using, uh, different tabs as a type of peer review between varied iterations. Sounds like that's what you're doing. Topai Ajayiho says, Your thoughts on agentic AI have been very helpful. Um, this week at the Blue Chip Data and AI Summit, he was a panelist on the future of work, and he did share the importance of focusing on the pain value against the tools, which are moving at a really fast pace. So there is some pain that's involved with keeping up.

Thoughts on that very quickly.

I feel the pain all all day long, so so it's, uh, I I completely agree.

Okay, Aaron, we have just a couple of minutes. Uh, what advice do you have for CIOs in relation to AI adoption?

I think it's the most exciting moment, um, in history to be a to be a CIO. There's there's so much, uh, so much change within technology. So you have, you know, if if you're even remotely curious as a as an IT leader in the IT organization, there's just a a a an incredible tsunami of of new things to play with and and technology to to try out. Uh, so I think that that's very exciting. I think, actually, the importance of the the CIO role, you know, rises dramatically, um, because this is again kind of back to a a prior point.

Like, agents are are maybe the most technical solution that has ever been deployed to non-technical people. Um, it it, you know, you're deploying non-deterministic intelligence, uh, into the hands of of every knowledge worker. Uh, that that it will, it will run wild and and, you know, grab the wrong data and produce the wrong report, or produce the right report and produce the right data and and generate the right software code. But that's all determined by your technology architecture.

It's all determined by how you've deployed these agents. It's all determined how you trained your users to use them. So I think the role of the CIO becomes, uh, substantially more important. Um, you're you're effectively providing work to your organization. That's the first time in history where IT was responsible for for deploying the actual sort of real output of the organization, not just the tools that enable the output. So so this is a again, an incredible time to be in IT.

Aaron, very fast. An important question from Arsalan Khan on X. The right agent with the right permissions can do great things, but what about agent deployments without any permissions and giving access to all data? How much knowledge is too little to deploy agents?

Because of our customer base and because of of how we've kind of grown up as a platform, um, we we see the danger in too much information, um, in in the hands of either the wrong people or the wrong agents. And so I I I, you know, it's j- this is this is definitely a phenomenon where you have to make sure that the right only the right context gets to the agent. You can't actually just shove as much context as possible to the agent.

First of all, that'll be a security nightmare. And the second of all, it'll actually still get the wrong answer because it'll it'll have too much things to put its attention on. So, uh, the right context at the right time, with the right guardrails is is still a critical problem for agents to work with.

Aaron, we're out of time. Give us the one minute sales pitch on Box.

Well, our job is to hopefully help customers solve many of these these problems with their unstructured data. So I think I think the reason why we're so excited and the reason I'm so passionate about AI is, uh, it's the first time in history where we can actually tap into all of this data that we have in our organizations. And so at Box, we're we're trying to build the leading AI platform to help companies tap into all of this unstructured data and knowledge, uh, and then work with your entire agentic ecosystem.

So I appreciate you having me on and and these are the topics that we get really excited by.

Aaron Levie, CEO of Box, thank you so much for coming back. This is your sixth time on CXO Talk and and I'm very grateful to you. It's been a fascinating discussion.

Thank you.

And everybody, thank you for watching, especially you guys who ask such great questions. You guys are incredible. Before you go, subscribe to the CXO Talk newsletter. Go to cxo talk dot com and we'll see you again next time. We have incredible shows coming up. Take care, everyone.