Bret Taylor of Sierra on AI agents, outcome-based pricing, and the OpenAI board

Bret Taylor is the ultimate Silicon Valley veteran. He was one of the creators of Google Maps, invented the like button, was co-CEO of Salesforce. He pushed through Elon's acquisition of Twitter when he was on the Twitter board. He's now the chairman of the OpenAI board. And his day job is founder and CEO of Sierra, which is bringing AI to customer service. He's one of the smartest people I know on the topic of how AI is changing established companies.

Cheers. Cheers. So, most important question, have you installed OpenClaw on your work laptop? I have not. Have you played with OpenClaw?

I have played with OpenClaw. I haven't bought a Mac Mini. You can put these things in virtual sandboxes for less money. It's really interesting. I mean, it's very compelling. It's probably the first, I wouldn't have predicted the first kind of broad, I don't know if consumer is exactly accurate, but maybe a hobbyist use of AI would have been this kind of semi-rogue open source project that goes through three name changes in three days. Yes. And I love it.

I love everything about the chaos of it, just because people in our circles have been talking about AI agents for consumer use and all these fancy computer-using agents, and instead you're chatting over WhatsApp with a thing on a Mac mini that is mildly unhinged and insecure. It's just fascinating. The whole thing is fascinating.

But isn't that, okay, the thing that seems to me is funny is if you look at the landscape, still in 2026, if you open a new Gemini chat or if you open a new chat, it's basically a blank slate. There's no memory. And then, I mean, people talk about the WhatsApp and Telegram integrations and things like that. But it feels to me a big part of the value is not only can it do stuff proactively, but it has memory.

But the way it has memory is this like super janky, it's like the movie Memento. It like writes things to a markdown file. And it's just like writing the things to remember. And the compaction is kind of buggy. Like it doesn't always write down the exact right things to remember and stuff like that. But isn't it funny that like you can get super polished mainstream consumer apps that have no memory at all. or this like wildly insecure three name changes project that kind of almost remembers things by scribbling notes in the margin.

And like that is the state of consumer AI.

I have a probably not very thoughtful, but kind of technical theory on this. So coding agents have gone through a transformation over the past four months. Like the difference between if we were here in October versus now our conversation about the future of software engineering would be materially different And how often can you say that about a technology? And people always, in my circles anyway, you look at a coding engine and you extrapolate to other domains.

You're like, could all digital tasks be like this? And the answer is obviously yes over some period of time. But it's really interesting because I think sometimes the hard part of engineering is in the details. And code repos have very specific qualities. One is all the context is in one place, in files that are largely textual, not binary. And for most broad information tasks, that's not true. You're making, like when you're writing your annual letter, my guess is the sources of information were in so many different systems, data warehouses.

And so it's not like impossible for an agent to use those things. But the idea that you can like straight line from coding agents to writing the Stripe annual letter, I don't totally buy. Yep. And then similarly, when the agent's actually performing work on a code base, there's feedback. There's compiler errors. There's often unit tests. There's integration tests. There's the history of every change that we made in a really formal format, along with code reviews. And so you can actually, it's almost designed for a robot.

And you can self-reflect. Maybe we as engineers are sort of like have always modeled ourselves after robots. And now we can actually fully realize that vision. So what's interesting about it is like the idea that it wrote a markdown file for memory, I think is maybe more significant than a hack. I actually think to some degree. Turning your life into code kind of. Yeah, it's like you almost want to put all everything in a file system that sort of looks like source code.

Not because that's the only way these agents can work, but actually it's quite an efficient way to get a mix of context and random access memory. if you think about like a vector database, it's more random access. You have to know what to look for. But actually, that's not how real memory works. There's a mix of it. So you're loading a markdown file. And as you said, compaction, all these things matter. But the messiness of it actually probably produces a more useful agent than a lot of the fancier things.

And I use memory in chat GPT, and I love it. But I actually think this idea that there's a directory of just everything you've ever done is actually maybe more useful to an AI than people think. And actually, if you follow over the past couple months, just this emergence of harness engineering, where you're building the harness around an agent to do work, I wonder if in the short term, it might just be one of those idiosyncrasies of history, like mimicking a code base is actually the best way to make a general purpose agent work.

And maybe over time we'll get fancier than that, But it's actually like a relatively efficient harness for an agent. So anyway, maybe that's why. Maybe that's why. Yeah, and it's very terminal centric.

And yeah, it's kind of backwards compatible. You can use grep. Yeah. You don't need to make some vector database. And the AIs really know how to use all the Unix tools. And so you get a lot of lift from that.

That's exactly right. I mean, software engineers were notorious for making tools for ourselves first. So then we just almost like bend every other domain in digital towards that domain. But the reason I brought up things like unit tests, integration tests, OpenAI did this blog post. I can't remember the engineer did the post, but on Harness Engineering. And one of the more interesting parts of it was documentation. So rather than just having a single agent's markdown file, they had a directory of essentially the entire product, the architecture, and they're sort of filling this out over time.

And agent's markdown became sort of pointers to it. But my hypothesis, having used Codex a lot, I wonder if the output of a session where you make a change to Stripe's products should be a documentation artifact in addition to code, where the documentation artifact is actually what the product manager version of John and the code was the engineer version of John, where there's a lot in the code that is more transient. You know, you might be fine to delete that.

What was the intention? What was the PRD? like what was the customer problem is actually the more durable asset. And I wrote this on X and one of the funniest comments would be like, it would be the greatest irony if, you know, software engineering agents made all of us just write documentation the whole time, just because notoriously every good engineer hates writing documentation. Now that's our job. But I don't know, it resonated with me.

How much are you AI code? Like you're a very prolific engineer in the old fashioned handspun way of writing code. And so how has that changed? The spoke artisanal code. Exactly. Yeah. Pour over.

Yeah. Pour over code. That's a really fun way to use that. I am trying to get to a world where I'm not writing code. It's hard emotionally, if that makes any sense. I have a hard time not caring. I don't care about the assembly language produced by the compiler. So I... So why should you care about the code? Why should I care about the code? You know, I care about correctness. I care about robustness. And I think I know intellectually I don't need to look at how the compiler unrolled this loop to verify its elegance and correctness.

Yet somehow I feel that way about code. And I'm not saying the code doesn't matter, but I've been trying to force myself to not care because I feel like I won't be like a self-actualized software engineer in the future. if I'm too precious about that artifact, which used to be so central to me. Right now, writing markdown files, like maybe that's fine. It feels somewhat like a local maximum and maybe we'll just be like, oh, of course it's markdown as hell.

You know, we work with machines. If you think about what a compiler does, there's this interesting mix of like formality and informality. And if you've used like Python versus Rust, sort of different ends of the spectrum. Now that you're not writing the code, I really wonder what that programming system should feel like and look like. And I don't mind chatting with Codex, that's fine. But I also think, as you imagine, all the tests that you care about, all the, it showing you demos and mock-ups, and I wonder sort of what the future integrated development environment, for lack of a better term, will be in that world.

So what I'm trying to do is force myself to not be emotionally attached to the code, which is very hard for me because that was my entire life before. I was proud of the elegance of the code that I wrote. but if I still care about the craftsmanship, what do I want? And I haven't quite visualized it yet.

It feels to me like a very interesting time in agentic engineering because you were talking about this domain of harness engineering and people having skills and MCP and everything like that. It's always interesting when not only is the leading product in a category changing, We're just figuring out what the categories are that, you know, we need things for MCP or, you know, skills or stuff like that. And it's all very fast moving. And that just feels to me like a very interesting time where clearly a new way of engineering is shaking out.

And 2026 is clearly not the final word.

Absolutely. And in fact, I'm growing more skeptical of MCP as like a meaningful part of the future. Not, it's fine as a protocol. But it's interesting, going back to your joke around OpenClaw just writing a big markdown file. I think it works better than a bunch of MCP servers. But going back to the point of every AI agent knows how to use crap, but knows how to use all these things. I feel like this view of a multi-agent world was you have all these agents that do tasks that are fraud detection.

You know, another one over here for, you know, personalization. And then you make a super agent, it does all these things, and it looks really good on a whiteboard, like most elegant looking but completely nonsensical architectures. And then you realize if you just, you know, imagine you anthropomorphize like the Stripe experience and you're like the checkout concierge. What information do you need to have no a priority to actually make that like a humane experience? And what ends up happening is multi-agent systems is you stuff all the context in the sub agents and the one on top has no ability to actually like not sound robotic.

And then in contrast, you look at something like OpenClaw, it's just a bunch of markdown files and it's the memory kind of feels right, even though it's a little bit kludgy. And similarly, if you go back to my arguments about a source control, you know, like a repo, it has so much context. So it's not like you just have the myopic view of the file you're editing. It really has some expansiveness. My sense is we're making true agents over time the way we think about context and how that context is sort of shared so that the agent that's orchestrated it actually understands what's behind all these APIs and why and the history.

We'll maybe look a little bit more like OpenClaw and less like MCP over time. And I think these agents need a lot more context than what MCP affords.

Yes. One thing we've noticed is there's a bit of a what's old is new again phenomenon where with this agent of commerce stuff that's happening, we actually built the APIs for this like 10 years ago as part of, do you remember that move of social shopping that was like for a while like buying on Instagram, buying on Twitter? Yeah. And it didn't quite happen for a few different reasons at the time. But the concepts are very similar that you want some action at a distance you want to be able to go kind of manipulate stuff um off-site and similarly you know i think patrick has wanted for the longest time in stripe is uh the ability to just ssh

into your stripe account you know what do you mean it's like a very ergonomic way for developers to work is you just be able to like log into your stripe account and you have a command line there yeah and then you can like list out all your pay or you can like tail the payments log or you want to tail and pipe and crap yeah exactly and of course now we're building that because it's like much more relevant in the agentic world but i find somehow yeah all the agentic stuff is also bringing back i don't know if you have this experience as well it's bringing back a lot of ideas that you might have

had before well it is because to some degree the elegance of unix which has sort of been the basis of of why everyone wants ssh and like the curl command that was sort of famously on the stripe homepage, for people who got it, it was remarkable because you could have all these tools that did something well, that was small and useful, and you could chain them together to make something great. I actually, I wonder in the future, we've talked a lot about this, you know, if you look at the canonical software as a service application like Stripe's console, and obviously you have your, what's the consumer see, but like the configuration that a Stripe customer will log into, you would

have a web app, and that's like the forms and fields and buttons and graphs. And then you have the API. And it was typically like a REST API or a GraphQL API, and you could do stuff with it. And this is how computers talked, and this is how humans used it. I wonder if the web application of the future will actually be, certainly you'll want a web app for the rare human who wants to sign in, but will you have an agent harness?

And what I mean by that is something more than the APIs, but just like if you think about the harness that you provide in a code base, the skills, the documentation, the roles. Imagine the person who's the greatest Stripe expert who knows how to extract the most value from their Stripe account. That's the harness, not the API. That's the button you click. And will that be an endpoint on Stripe .com and so that your agent knows how to just get the most value from Stripe.

And I imagine you don't care if your merchants log in. What you want them to do is drive value for themselves, drive GMV, drive payments. And so I'm really excited about that because an API is great. APIs are awesome. But a harness is basically like, here's the instruction manual for all the Unix commands that power Stripe. That's very interesting.

Yes, and I think if you look at the shape of a lot of APIs that services have, and I think Stripe's API coverage is probably more complete than most, but ultimately it is a way to manipulate some of the highest value business objects in the thing. Whereas actually what you want is one, all of the data to be browsable in some kind of identically accessible or textual way. and then all of the actions to be able to be taken by agents.

And it turns out there's a lot of switches in the dashboard. There's no API for it, and we are all as an industry collectively discovering that.

And it might be easier. Imagine being a product manager in the future. You just need to add the switch to the dashboard. You're like, yeah, it looks like a Russian submarine to switch this, but who cares, right? Agents can handle it. And as long as the harness describes when to use that switch, you know it's easier than UI design in some ways and that's fascinating to me.

But one funny point Dario made is it's not clear will the there's a race between people getting their stuff accessible via agents and just can desktop computer use getting better and so it's actually not clear will the approach be Stripe builds way more APIs and that's how you know your agents manipulate the Stripe account, or you just give your agents access to Chrome and a login.

Well, so actually, I'll give a funny story here. So, Sierra, my company.

Sorry, we'll get to Sierra.

No, no, it's fine. But there's a real funny story here. So, Sierra powers a lot of healthcare companies. So, like, on the healthcare payer side, health insurance. Well known to their API quality. Well, so, first of all, they're actually pretty sophisticated engineers in these companies. I I really enjoy working with them. So you end up with Cigna, Blue Cross Blue Shield on the healthcare payers, right, insurance. Then you have healthcare providers like Sutter Health that we work with.

Then you have revenue cycle management. So like R1 and revenue cycle management basically help providers get paid by the insurance companies. And then you have a lot of other people in the middle of pharmacies, PBMs, and they all call each other. So like healthcare provider has to call a payer because a procedure happened and they have to get paid. So we have payers with AI agents that pick up the phone. And we have providers that have AI agents that pick up the phone and make phone calls.

We have revenue cycle management companies that work to make outbound calls to do it. We've already had... Do they switch to the agent language? They don't. We've done English over the publicly switched telephone networks. You know, TCP IP and English over PSTN. And it goes, I mean, it sort of reinforces, I guess, Dario's point, which is you can engineer all these fancy protocols, but the rails that are already there already exist. English is spoken by all AI agents, and the publicly switched telephone network has been around for 100 years, and it all works, which is fascinating.

So you have all these fancy MCP things, and we're doing like English over PSTN. So on one hand, I think I actually agree with the principle that one of the powerful parts about AI with its ability to do text, do audio, and do, you know, I'll say, I'm not sure how you qualify computer use, but you can call it a form of image recognition and manipulation. Certainly that's useful because you get to the point where you don't need to, like, fully finish the last mile to get value.

The thing I'd say, though, going back to your, like, talk about all the actions and just they're not all in the product, all the APIs don't exist, these visual interfaces were designed for us. So think of, I mean, Stripe is sort of, I think, famously was one of the few enterprise software companies with good design for a long time. And I like the Stripe dashboard is really elegant, right? And most enterprise software, you can't say that about their dashboard.

I don't think the ideal agent, like agent harness will be that elegant because it's optimized for something else. It's optimized for the context that you need to perform complex multi-step procedures on behalf of a person. And, you know, my guess is it's just very different. And I think, seeing the way you write a harness for a software or a coding agent is just so different than the way you do iDesign. So I'm certain that it's great that you can click around a green screen or whatever, click around a green screen because it's an oxymoron, but type around a green screen or click around sort of a legacy on-premises enterprise software system.

I think these harnesses will be really good. And I wonder if, is there a world two years from now where Stripe's ability to work with the agent that manages commerce for a direct consumer e-commerce company, that would be one of the ways you're evaluated. And in fact, if you're, for lack of a better word, harness is not compatible with the way their agents work, that's actually like you're not compatible with them. And I'm not sure that's right.

So I think it's great that these things are backwards compatible. It's great that our agents have spoken over the telephone already in English. That's really funny. But I don't think it's like the long-term future because there's so much value that you can provide. Put another way, the agents using a sophisticated application harness can just do a lot more and do a lot more like with higher fidelity.

Yes. Well, we should get to Sierra because you see a lot of real-world AI adoption. And so maybe start by grounding us. What is Sierra? The business has scaled very quickly. So one of the latest metrics that you can share because they keep changing from month to month. Yeah.

So Sierra, we help companies make AI agents for their customer experience. So if you have a big phone line, these agents can replace your IVR system and just pick up the phone. If you have a digital chat system, an AI agent can pick it up. You don't need to wait on hold. These agents can not only answer questions but take action on your behalf. We work with, you know, healthcare companies like Cigna. We just did a great case study with SoFi.

And I'm really proud that we raised their net promoter score by 33 points just because it's just so delightful. It's really fun to see, you know, sort of all these different brands across a wide range of industries get so much value from their agent. With a leader in the space, like just you talked about the metrics, we reached $100 million in ARR in seven quarters, $150 in eight quarters. We're, I think, around $165 million now, one month into our, you know, next quarter.

So growing really rapidly and really proud of the momentum that we have.

That's super cool. What is the typical adoption? Are people using it for email chat support because that's the easiest modality? Do they adopt it for everything, including phone and stuff like that?

It's changed a lot over the past two years. But I'll say the median customer, and they'll describe some interesting outliers, which I hope are sort of gumsens of the future. So most will start with one channel and a few use cases. So, you know, at a lot of healthcare companies, phone remains the dominant channel. So say, hey, for a few types of phone calls, let's have the AI agent take them and see how it does. Do people like it?

Are people comfortable with it? Does it lower our cost? Does it raise whatever, you know, metrics? Usually it's customer satisfaction. Yeah, yeah. And does it work more effectively? So, for example, like for a car insurance company, it'll be like first notice of loss. You know, I got an offender bender, you know. And that would be the typical way you start. For a lot of more digitally native companies, they'll start with chat and kind of similar. But almost all of our clients will do both.

So SiriusXM, if you call them on the phone, their AI agent, Harmony, which I love that name for SiriusXM, will pick up the phone. And if you go to their homepage and you see the chat, that's also the same agent. So the neat part is, I think it's pretty neat because you have literally all of your, I'll say, customer experience team or whatever you might call it at your company. they can spend all their time on one thing.

And it actually works over WhatsApp. It can work online. It can work in your website, work with mobile app. It can pick up the phone. That's a pretty big change. A lot of our clients, when we start working with them, they'll have like a digital team and a call center team and all these different teams. And we've kind of gotten to the point because we've digitized the last remaining analog channel, which is the telephone. Those are all unified.

When I started my sort of glimpse of the future, we have a few like really ambitious customers like Rocket Mortgage, great Detroit company. They own Redfin. They bought a mortgage services company called Mr. Cooper. If you go to Redfin, you can search for a home using an AI agent. If you go to rocket .com, you can originate a mortgage with an AI agent and you can service that mortgage.

It becomes product usage rather than just customer service.

And really end-to-end, sales, service. And I think that's really exciting. I mean, our whole view is that if we're in 1994 and you were doing Cheeky Pint about this internet phenomenon. I don't mean a fan. I was a bit

young, but yeah.

Yeah, I was. I want to have my Nirvana shirt on. You know, we would be talking about like, look, this is like your digital front door. Or maybe we wouldn't have the pressures to say that. On the information superhighway. On the information superhighway. And I think the same is true of most companies, AI agent singular. There are lots of agents, but the one with your brand at the top that your customers interact with is special, and that's the one we're trying to power for companies.

So you think this, like what customer is built on Sierra, your aspiration is that it just becomes sometimes the primary way people deal with the company.

I think a company's AI agent will be the vast majority of their digital interactions. Yeah. And I think digital has come to include the telephone. And that's sort of a big shift because we think of that differently. Yes. And that's a huge change just because the bigger shift, so customer service, which is one big part of what we do, but not the only thing we do, is traditionally been thought of as a cost center because it's really expensive.

So I'm sure you have people answering the phone for your clients. And depending on where they're located and how well-trained they have to be, like how simple is the case, it can cost $10, $20. It can be much less if it's a more simple case. And, you know, you have some customers who pay you millions of dollars, and you'll answer the phone anytime they call, and you might have one that has not even started monetizing yet.

and you might want to call them, but there's a limit to literally how much you can afford to talk to that person and still have a profitable business. I always joke, it's probably easier for you and me to call Sundar than to get Google customer service on the phone. It's very hard to get Google customer service on the phone. But it's not because they don't like you. It's just if you think about the average revenue per user of Google, they literally just can't afford to do it.

So now if you take that $10 or $20 phone call and you make it $0 .10 or $0 .20, and over time, $0 .01 to $0 .02, all of a sudden not only can you afford to provide a great customer experience to more people, even less profitable customers or in lower margin businesses, which I think is very exciting. So it's like not just doing what you did before, but new. You can provide better customer service. You really can.

And then just think about running like a subscription business where you care as much about customer acquisition, you care a lot about churn because that's how your lifetime value equation works. and you think about, okay, if I had a budget of how much I spent on service and now I can do 100 conversations more than I could before, can I actually reduce my turn rate? Can I improve lifetime value? And then the interesting thing is then you realize that, wow, all of my competitors have access to the same technology.

And then you're saying, okay, what are my competitors going to do to actually take my customers away from me? And then that's when you start to get things like, you know, The ATM machine didn't actually reduce bank branches because some bank had the great idea of, I'm going to put different people in this branch. It'll generate revenue. And all of a sudden, it wasn't job displacement, but something completely different. So I think the exciting part in our world is you're taking something that's just so, so, so expensive that people literally hid their phone numbers.

So people couldn't call them. And you're making it inexpensive and delightful. And the thing I'm excited about is the second and third order effects are going to be really interesting. and very hard to predict. And that's pretty exciting.

I want to come back to this idea of the agent as the UI, because I find it really interesting. Like we talked about this a bit in our annual letter in the context of agentic commerce, where again, I think people are trying to pitch too much of the end state of like, you know, fully autonomous, you know, the robots just choosing few. And the point we always make is like, let's just start with not having to fill out the web form.

Like no one likes filling out forms on the internet. Speak for yourself. I wonder just, will using websites have been actually a bit of a, like the fax machine, you know, we used to hum emails over the telephone lines as a way of transmitting information. Or like, I wasn't working for this, but like there was an era of like voicemail memos where you're in the working world for that.

Some people still do this where they, you know, do the voice.

But like companies will like blast a voicemail memo to like employees at the company. And like, that's a way of distributing information. And all these things are like very moment in time. and maybe navigating websites and filling out forms was like a bit of a moment in time. Is that how you see things playing out?

I don't know. I mean, it's interesting because if you look at the past few iterations of technology, you had the PC revolution, then you had the internet and the browser, then the smartphone came out and the tablet. And I was more optimistic about tablets than sort of the way the world turned out. You know, I see more tablets on airplanes, but like I don't, you know, I'm guessing if I walked around Stripe, I would see very few tablets out.

And similarly, there's more smartphones than people, but there's still about 2 billion PCs in the world. And I think it peaked some number of years ago, but it hasn't gone down as far as I know. And I haven't tracked this. That's interesting, right? We sort of added to our digital world, but I think perhaps the more interesting metric is for like you and me, what percentage of emails were sent through each device? And certainly from, you know, 2010, to 2020, most of the world might have transitioned from like percentage of email on desktop to smartphone, you know, significantly.

And so it's almost like a market share of digital interactions, which I think is a really interesting way to think about it. And certainly as you think of like Stripe's business, like where does commerce originate? And you saw that move to mobile, but it doesn't mean that people, it's actually a very big, you wouldn't want to eliminate the PC e-commerce business, that would actually be catastrophically bad. And so then you look at AI agents, and I believe most businesses, it will be their primary digital interface.

And it's because it works over WhatsApp and it works over the phone. If smart speakers make a comeback, they'll work over smart speakers. Which they may now. Totally.

Smart speakers were just too early.

Yeah, well, it's like ask for the weather just turns out to be like not the biggest market in the world. But now...

Set a timer. Set a timer. Yeah.

I mean, it's amazing they made that much money off a timer setting speaker. And so it is very future-proofed because it's fundamentally conversational. But maybe it's like going from punch cards, mice and keyboards, touchscreens, now voice and chat and probably 3D immersive at some point. Does it just sort of add and make the other ones less important is probably the way I think about it. I do wonder if we'll see the end of the smartphone at some point.

It doesn't seem anywhere close to right now. But it is interesting. I mean, I think most people don't love how much we're sort of addicted to staring at this glowing screen. On the other hand, you can't talk to TikTok. You know, it's fundamentally visual. But I wonder if there's a world where you could actually be really productive without such an invasive device on your body. Yes. And if that's the case, can it offer an opportunity to sort of like unwedge some of the addictive properties of these technologies and get a lot of the benefits from it?

You know, because like, at least for me, like, I think all of us are so connected. You sort of end up like, I'm going to check my email. And then like, you're like, where have I been for the past hour? And the fact that we actually have technology that affords that kind of innovation now, I think that's quite interesting. So I don't know what the future is, but I'm very excited for it. I know it sounds really cheesy, but we've now changed the ingredients available, and we have a lot more recipes we can cook, and I think that's very exciting.

Yeah. I agree that I'm excited for not having to look at the screen for as many things, for a variety of reasons. When a customer installs Sierra, I know there's a significant customer satisfaction component as well as cost, but I'm curious what kind of cost difference does it make, and maybe relatedly, when a customer is fully deployed, what kind of mix do they see between queries fully resolved agentically, things that end up having a human who is presumably somewhat AI-assisted, but just what does a normal equilibrium look like there?

Yeah, it turns out most of our clients have pretty different priorities. So some are very focused on cost savings, and you can automate very, very high percentages of your cases. A lot of, there's a company called Ramp that's a really impressive tech firm.

We had Eric just here. Oh, that's great.

Well, they're automating 90 % of their cases. They're really sophisticated, though, because they're basically getting in front of cases before they escalate. But I think it's kind of an example of just a really fantastic company implementing really well. And you can see anywhere between 70%, 90%, which is really incredible. The interesting thing, though, is there's counterintuitive effects to it. The cases that do make its way to your customer service team can end up more complex, sort of by definition.

So what's called average handle time will actually go up. And we heard from one of our clients that actually their satisfaction of their call center agents went way up, too, because it turns out it's way more fulfilling to solve a hard problem. Like, have you tried plugging it out or plugging it back in again? Exactly. The other thing when we had one retailer whose volume, total volume went up almost as much as they saved from the AI.

It was a form of that. So, you know, if you've used a chatbot from three years ago, they were really annoying. Like three years ago, if you said, like, do you like chatbots? There was like zero people would say yes.

It's so funny that there was a Silicon Valley wave of hype around chatbots. It was even earlier than that. It was like 2018. It was like pre-LLMs, pre-transformers. Yeah.

And they were just like multiple choice machines or something. It was just the worst products of all time. And so replacing it with something that was like a delightful way. People are like, I'm going to talk to this thing a lot more. So they ended up keeping their cost didn't really go down, but their volume of customer conversations went up, you know, two or three X. Yes. And the CEO was incredibly happy about it. They were like, I've just, we're now actually listening to our clients.

So it sounds funny, but it's a little bit of a choice how much you want to drive cost savings with AI versus other metrics. Most of our clients are interested in the top line metrics. And so if you could save $10 or save $1 and improve your net promoter score and competitive positioning by a meaningful amount, everyone in the world would choose the latter. So that's the interesting thing going on right now because, again, going back to about, it was 1994 and we're hawking websites on this podcast.

I think if we were to go to a major bank and say, if you launch a website, you're going to have a competitive advantage against every other bank. With the benefit of hindsight, that would have been overpromising. The correct thing to say was if you don't want to launch a website. You have to have a website. And so this technology is broadly available. And so as a consequence, you can't just sort of like launch in all parts of AI, not just Arbus.

You can't just launch AI, absorb the cost savings, pass it on to shareholders unless you have a monopoly. Yes.

In most businesses, it's a consumer surplus.

Exactly. So you're either going to lower prices, but I think that's why it's an overused analogy, but the ATM bank branch thing is really interesting because if every single company in an industry has access to technology, I would say it's an imperative, not a competitive advantage. And the more interesting, I would say, board discussion is when everyone adopts the obvious things, customer experience, customer service, software engineering, legal, just pick the ones where there's solutions, off the shelf solutions available now, what will the industry look like?

And my guess is you could ask how GPT think. And my guess is there's some really interesting second order effects. And when you have competitive markets, you're going to end up investing, lowering prices, whatever it may be. And that's the thing I don't think it's talked about enough. And I actually think that we just, it happens with every technology change. You project it through the lens of what you're doing today, and you don't take into effect.

It's like a multiplayer game that we're all in right now. Yes. And that's fascinating to me. And so the change is disruptive, but I think it's going to be like, I'm very excited for the next few years as the world absorbs the technology. we start getting to some of those second and maybe even third order effects.

What is the most impressive AI adoption or kind of AI native behavior you've seen from a client?

Oh, that's a really good question. I'll probably say Rocket, where we have a really great relationship. I think Varun is their CEO, Sean Mahutra is their CTO. Two people who really are, I would say, not only just curious about AI, but very interested in transforming the home ownership experience with AI. And I don't know when you got your first mortgage, but it's very intimidating. And it's not a modern process. It's not a modern process. They literally call it mortgage folders for a reason.

It used to be a folder. And for me, it's an example of a company trying to transform an industry. And the reason I brought it up in the context of our last question is it's not just saying, how can we take AI to do this? But it's like, if you were to think about the homeowner experience from searching for a home on Redfin all the way through servicing it, and you had AI available, what would the ideal experience be like?

And so that's really interesting to see Rocket with their acquisition strategy too, kind of like integrate that experience. And that's what I'm excited about. I think there's an opportunity for CEOs and like industries like that to have a bold vision of like what the future could be. And going back to my point, imperative, not competitive advantage, it is a competitive advantage right now. So if you imagine, I haven't tracked the market share of all the big U .S.

telcos, but if you look at T-Mobile, Verizon, AT &T, and you tracked it over the past 10 years, you end up with surges in market share growth. The iPhone came out. You ended up with 5G. And you end up with these things where, but it's my impression of the industry you end up with these moments that drive market share, and then it ends up at an equilibrium. So what's interesting about it is the iPhone moment for telecommunications companies like SoftBank in Japan.

This is the moment where perhaps if you have a competitive equilibrium, you can absorb this technology, use it, and you'll have this window where you can actually shuffle the deck.

It's a technology that shakes the competitive equilibrium. Yes, exactly right. I definitely noticed that. Yeah. So you talked about how coding is such a domain that is suitable to AI because all of the context you're working with exists in the repo. It is in text. It's kind of neatly organized to be executed and read by humans. And so there's kind of a good bounce there. the problems that customer service agents are not of that character. And so how do you actually smush everything into a format where your AI agent can answer it?

Yeah, we spend a lot of time thinking about that. To some degree, one of our engineers called it almost like we're creating a domain-specific language for specifying customer experience. You know, like what is the mechanism of specifying it? We use this metaphor we call journeys, which is, you know, what is a customer journey end to end? And what does the agent need to be successful in that journey? What tools does it need to access? What information does it need to access?

And you can, if you think about the capabilities of an agent like skills and a coding agent, you'll add different capabilities over time as the customer is talking to you. The key thing that's been a breakthrough that is probably not surprising to the technologists listening to this, but has been a huge difference between those crappy chatbots of four years ago is the reasoning capabilities. You know, I think the... Well, we had one client who had acquired three companies, and they had three identity systems, three CRM systems, three of everything.

And so they had this big IT project where they were going to unify all those systems. But I was like, why don't you just have the agent like go in all three of them and just think. And they're like, well, what if there's duplicate data? What if the data conflicts? And they're like, you know, that's going to fool. Sensor fusion. And I was like, well, what does your person, what does like the person do? Like, well, they kind of think about it.

And I was like, oh, let's just do that. And that's the interesting thing about these AI agents is they actually, the basic humane, basic reasoning, not superhuman ASI, turns out to be the huge breakthrough in customer experience. The other interesting thing is the innate knowledge of the LLM. You don't want an AI agent to hallucinate, obviously. But Sonos is one of our clients. And do you have a Sonos speaker at your house? I have had, yeah, yeah.

If a Sonos speaker ever breaks, it's never the speaker. It's always Wi-Fi. That's what I've learned. And it's always true of me, too, right? There's always some Wi-Fi. You know, if you wanted to make an AI agent to help you with your Sonos speaker, like you obviously can give it all the manuals for the speakers all the technical specifications you can give it the device telemetry all the stuff you need do you really need to give it the history of wi-fi well now it turns out like large language models have encountered every possible wi-fi problem so like why does the sonos ai work so effectively well it knows a lot about wi-fi in addition to

all the sonos things and if you look for any given ai agent yes all of the like it turns out being trained on all of human knowledge is actually useful as a starting point for a lot of tasks. And I think that's been the big breakthrough. So how do you give it all of its knowledge? Well, first, we've built, I think, the best platform in the market to do so, where you can really narrow the guardrails for regulated conversations, widen them for less regulated conversations.

But the fact it starts with knowledge of obscure Wi-Fi idiosyncrasies turns out to be the greatest

breakthrough of all time. Have you had the opposite problem where there's a customer whose problem domains mostly don't exist on the public internet. You know, it's like we provide the drill bits used in, you know, deep sea oil drilling. And it just turns out there's nothing on Reddit about that.

Yeah, so 100%. And, you know, we work with this like medical device company and it's a deep cut of human knowledge. You know, and you can train it all on that. In fact, we do a lot. One of the things you want to be really careful about if you have a really well-known brand, We work with, I want to say, a third of our clients have over $10 billion in revenue. Over half have over a billion revenue.

So most of our clients are actually quite well-known. So one of the challenges when you're offering either sales or service or customer experience to a really well-known brand is it's harder to ground it. It's actually easier when the Internet has never heard of you and you want to make a well-grounded agent. It's actually pretty easy because there's no temptation from the LLMs to go off script. So actually, ironically, the harder challenge is when it's a very well-known brand, it's like, no, I got this.

I'm like, no, you don't. You got to go look it up. You know, that's actually a harder problem.

And so how do you force the LLMs, like mechanically, how do you force them to not, you know, answer off the top of their head, but actually look it up?

So we use, we call it a constellation of models. So our platform, we call it Agent Studio, you essentially configure the goals and guardrails of a process. And goals and guardrails, not a sequence of steps, because you want agency, but you want guardrails around it. And within that, we'll use reasoning, but we use supervisor models to actually inspect that reasoning. And so if you were an AI agent in Sierra and you decided to go off script, like, I got this.

What would end up happening is a supervisor agent would observe your reasoning, say, I think John should have actually looked up the policy here and send it back with notes and say, actually, you're not allowed to make that decision. Here's the reasons why. Go redo that decision. It's a really effective technique. The way I think about it, which is a little simplistic, but I think basically right. If you imagine a reasoning system is right 90 % of the time, but has some either guardrail malfunction or hallucination 10 % of the time, it's obviously better than that.

And then you have a supervisor that's right 90 % of the time. If you chain them together, you get 99 % effectiveness. And so that methodology of layering reasoning and intelligence has been really effective. And in general, it sort of makes sense. you're basically layering compute, you're layering reasoning on top of it. What's neat about it, though, is we can sort of abstract that complexity from our clients. So, you know, they're sort of expressing the goals and guardrails, and we have all these evals and tests and all these other things.

We can find ways to make it more and more and more robust over time, but it doesn't require you to, you know, prompt engineer, write in all caps or whatever, like the hacks that people use to get these things to be conformant.

And you started in 22, 23?

We launched the company on February 13th, two years ago. So I guess we're like... Yeah, so our two-year birthday was like a couple weeks ago.

What I was thinking as you were saying that is, did you sort of co-evolve Chain of Thoughts and RL and some of these things that are now in the models, but did you have to build your own kind of janky version of them before they were in the models?

So, yes, and it's also a talk, which is the weird part about building a product right now and a company right now, because so much of what we write we plan to throw out later. Yes. And this is a very weird way to build a company. So Google's chain of thought paper, which preceded 01 and doing reinforcement learning on chains of thought, was out roughly when we started the company. It was an earlier paper and effectively provided sort of a substantive basis of why asking a model to explain its reasoning step by step produced more robustness.

So we use chain of thought all the time. And it was like a methodology we used. And then OpenAI very innovatively came up with the idea of we could do reinforcement learning on those chains of thought, which is where O1 came from, and then most labs are doing that now. So we'd throw out things all the time. You'd do it, and you're like, okay, the model just does this for us now. We work with a lot of financial services firms.

We work with one bank that has a large Hong Kong business, and they speak Cantonese. And like, okay, well, we need really good Cantonese voice support. And it turns out that that's really hard and there's not an obvious model that does that. So we spend all this time evaluating all these models. What certainty would you ascribe to like every voice model supporting Cantonese well in three years? 100%, 99 %? Pretty close. So we did all this work.

And in fact, you know, we, I think, have the best Cantonese support on the market. Great for us. And it's a huge selling point. And it's a technology that will certainly be commoditized in three years. So a lot of what we think about, I think, is going from essentially technology innovation now. I think a large part of why we work with the largest companies in the world is because our technology works. In three years, those same clients will work with us because we have the best product.

And I think if you look at the early marketing for early softwares of service companies, they'll explain why having multiple tenants in the same database is safe. And that was a huge part of their marketing. Nowadays, if you came and you marketed your product that way, people are like, what are you talking about? I don't care what database Stripe uses. I think we're just at this period where the technology is so immature. it's a very technology forward conversation just because it's like people are figuring it out.

Just like when Netscape's business was like monetized through a web server back a hundred years ago. And it will evolve from being a technology forward conversation to a product forward conversation. So the interesting part about building an applied AI company is you can't have the luxury of waiting for all the models to catch up with your aspirations. But you know they will. But you know they will. Yes. So you have to have the best technology and have to be comfortable with throwing it out.

Yes. And so it's a real momentum and pace of innovation game. Rather than thinking of this as like precious intellectual property, if that makes any sense.

It absolutely does. But isn't this organizationally hard where if I'm the head of Cantonese language as Sierra, my incentive, and then not disingenuously so, I'll notice all the corner cases where the models aren't that good in Cantonese. And obviously we saw this in prior tech waves, where the cloud adoption laggards were companies that had their own on-prem stuff and they had a million reasons half real half fake as to why cloud did not suit their business purposes but how do you avoid getting stuck in this mode of thinking where like oh well their chain of thought doesn't do what what we need is like the classic thing you would hear from someone within the organization

it's a huge shift i mean going back to the first thing we were talking about it's hard for me to not care about like the elegance of the source code which i think is an impediment to my fully realizing like being a software engineer in this new world i think teams that start to treat the code that they wrote that as precious yes that has been obviated by a general purpose ai model will fundamentally fall behind public markets deem

the software industry 20, 30 % less valuable than they did maybe three months back. A day ago. Yeah, exactly. Very recently. And, you know, the two sides of the debate are, one, you know, the valuations were based on what the businesses will do in 2030 or 2035, like far in the future. And just there's much more uncertainty there. And so this is deserved. And the counter argument is that it's still not the case that agentic software production is really going to build you a workday.

Indeed, Anthropic just installed workday, very famously. Where do you net out on is this a rational response or not?

I think it's rational, but I think it's a bit overblown at the same time. So I think it's rational just in the sense that there's probably been there hasn't been more uncertainty in this market ever. Yes. And so unless you have a strong thesis about an individual company, my guess is like, will these companies be less valuable 10 years from now than now? I think the answer is probably yes. Will that be true for every individual company?

I don't think that's true. And so if you're just thinking about a portfolio of investments, I think it's sort of an indictment of the sector more than it is an indictment of an individual company. I don't know if the value of these platforms was who could VibeCode in a weekend ever. Not that we knew what VibeCoding was. My point is, you know, everyone who's ever built a software-as-a -service application has had a Hacker News comment of, I could have coded this in a weekend.

Like every single one. Famously Dropbox. I'm sure you have as well. every single product I've ever made. It just happens. It's like a right of, in fact, if no one said that on your product, like I'm sorry. It's not relevant. Yeah, you're not relevant. It's not interesting. And obviously most of those comments were incorrect. But if you think about, you know, all the work you've done in compliance or the relationships you have with large financial services institutions, the work you do on fraud, the things under the surface that aren't, you know, the forms and fields in the web browser are actually incredibly valuable.

If you think about a large software company, You know, they'll have thousands of quota carrying account executives that represent sales capacity, which is basically a channel. And distribution turns out to be a very important part of software. There's social proof. There's the old saying, no one gets fired for buying IBM, which few people say right now. Though IBM is actually like doing really well under Arvin. You want to be maybe the first health care insurance company to adopt something.

There's another health care insurer who says, I want to be the fifth. I want other people to prove it. There's all these network effects around these businesses and scale and moats and sort of Silicon Valley speak around them. I think the big risk is where is value in the software industry and years from now? One risk is that more people will build than they do now versus buy because the marginal cost of writing software goes down.

I think that'll be true for some software, particularly developer platforms and things like that that are already being consumed and purchased by other engineers. Little libraries or, you know.

Things that already were part of the build versus buy calculus, it shifts the balance of power.

Absolutely. Yep. The other part of it is systems of record. So I think these systems of record have always been sort of the gravitational center of their relative solar systems. And it roughly breaks down by department. So ERP systems are associated with the finance department. And SAP and Oracle and Workday have ERP systems. And you have Adobe in the marketing department. And they had Salesforce in the sales department. And you had ServiceNow in the IT department.

And everything sort of rotated around them. And why? Well, first, their database was sort of truly the system of records. So every application that wanted to interact with the data in that had to, you essentially collect taxes from your ecosystem. And then similarly allowed each of those systems of record company to essentially have revenue expansion opportunities to go to adjacent areas where they're all sold to the same buyer and all of that. But the thing that's really interesting is AI agents are actually performing valuable labor.

Is the database and the system of record, does that continue to be the gravitational center of each of those workflows? So I'll just take marketing as an example. The database of, you know, your customers that you use to drive, you know, sending out an email blast on Black Friday has some value. but if you had an ai agent that drove like way higher like more leads for your sales team from that marketing blast you probably that's worth more to you than the the system of record itself similarly if you imagine uh you know uh i'll just take like a crm system and you think about the ai agent that's carving your territories if no one ever logs in to actually

do it manually all of those things have a lot of value and a lot more value than, you know, relatively speaking, than they did because they're actually performing the action. And so the real question to me is like, does it upend this, I would say something that's been true for 30 years, which is all the values in these systems of record. And the way I think about it is agents are to some degree, a system of record of a process of generating a lead or auditing your financials or reviewing a contract or whatever it might be.

And I don't think we've ever had a piece of software like that. And will those encoded, well-optimized processes start to have more value than the databases? I don't know that's the case. For example, if your ERP system is your company's ledger, that'll have a lot of value. But I wonder for all these others, and my theory is the closer you get to literally the database is the value, i .e. a ledger, the more durable it is.

the closer you get to be in a system of engagement uh the less terrible it is that's a very interesting

framing yeah and it kind of gets back to the point you're making about the company that was looking to standardize you know and not have three different drps and stuff like that and you're like why just try not doing that and i think maybe the consumer example of this is i think people have probably had the experience of you paste data into an lm to do something with it and you know like the formatting is all messed up and like, you know, the tabs and spaces don't come through and everything like that.

So it doesn't matter. LLM doesn't care. You can just like pay through whatever and it'll work with it. And so this idea that, as you say, if like the system of record is important because it's your general ledger and it matters to the auditors, that's one thing. But if it was a system of record in this kind of all your data in one place way, and because it was easier to build incremental software atop it, maybe that advantage is going away because the agents are fine plucking data from 10 different places.

That's roughly my view. But the bull case, I always mix up bear and bull. I need to spend more time on Wall Street. The bull case, though, is I think all these companies sort of have a right to win. They're all big. They still have sales capacity. They have all these advantages. But it's a race. You know, how fast will smaller companies build differentiated, scaled businesses before the incumbents go into this new world? But for a wide variety of well-documented reasons, you know, like disrupting their own business model, it's harder.

But I think your ask is, like, is it irrational? I don't think it's irrational. I think there's just more uncertainty now than there's ever been. And, you know, I think that's markets are telling you there's a lot of uncertainty. And that's why you see sort of people recede from the whole category, basically.

I feel like there's also a totally separate thing playing out here where for a long time, certain companies were criticized for not taking profitability that seriously. And at some level, there's just a return to normal valuation levels on like a fully loaded stock based comp baked in and everything basis. that's kind of independent of the AI thesis, but maybe just some return to, again, on a fully loaded gap basis, more normal valuations.

Well, essentially, if you look at a traditional software-as-a -service company, the way most people model it is you have annual recurring revenue, which is basically an annuity, and it should throw off that much cash every year. Then you have attrition, which is subtracting from the annuity, and then you have net new ARR, which is adding to the annuity. Your salespeople sell software to add to the ARR. You typically have an account management team or customer success team to keep churn down.

And you grow that annuity and you grow your headcount, often just a little bit ahead of that annuity because you need to grow a new business. And if that annuity is not an annuity, then that math really changes. It really changes. And so, you know, because the whole idea of software as a service is you can just, you know, slow down hiring and you become very profitable because the annuity starts starting off cash. That's been the thesis of every private equity firm who, you know, acquires slow growth software as a service companies.

If you don't assume that that revenue is going to be there two years or three years from now, your discounted cash flow analysis looks pretty different. And I don't think it's actually quite so dire in the timeframes that people think. But again, if you're asking, like, markets are, there's, you know, all those great quotes about Wayne and, I don't know. Voting Wayne machines. Voting Wayne machines. Yeah, I get it. Like, there's probably more safer sectors to invest in.

But I don't think it's an indictment of individual companies. That's my point. I actually think, you know, when we first met, I doubt either of us had an extremely positive view of the future of Microsoft. At the time, it felt like a previous generation company. Now you look at Azure, their open-air relationship, all these things, what an impressive turnaround. I think any one of these companies could do it. I think it's just, but it's more of an indictment of the market.

Yes, yes. I have a lot more questions. Would you like to know that, Guinness? Sure. Brett has been through a few platform shifts. And one thing he's been pretty consistent about is being mindful of the external forces that are shaping the ecosystem you're in. He talks a lot about building with the broader wave of AI agents in mind. Stripe Sessions is our way of helping builders see that wave up close. What's changing in the internet economy, what's actually working in production, and what the next era of software looks like when agents are running real commerce workflows.

It's not the usual conference fluff, it's insights into what the fastest moving companies are actually after. So if you want to experience the next chapter of the Internet Economy firsthand, join us at Stripe Sessions this April. Use the code Cheeky Pint for 50 % off a conference pass at sessions .stripe .com. You talked about business models. Are you guys usage-based or, yeah, how are you innovating on the business model front, or are you?

We are trying to. So we do outcomes-based pricing. So for a customer service context, that means if the AI agent resolves the case, no human intervention, there's a pre-negotiated sort of rate for that. And if we do have to escalate to a person that's free for sales, it would be a sales commission. And wherever possible there's a way to align our interests with our clients, we choose it. And I'm a huge believer in this.

I think the analogy of going from impression-based ads to CPC ads is apt. You know, I don't think any ad platform thinks like, man, think of all the impressions we're giving away for free. Because when you charge for something closer to a business value, it's actually more valuable. It's more efficient. It's a lot more efficient. And I think the idea, if an agent's outcome is measurable, it's a really compelling way to both for clients, obviously, because it's aligned with their business.

But it's also quite disruptive because most, I'll say, legacy software companies are not necessarily equipped to do it for a variety of reasons I'm happy to go into. But it's just a very disruptive model.

Yeah, there's kind of a few lenses you can have on it. One is that you get more alignment, like usage-based is more aligned than other ways of charging. And as you say, it's more efficient because you're incentivized to drive the right outcomes. people also make the analogies to, you know, it's almost like more correct for the labor substitution dynamics that you get, or just like because you have real inference costs, you're going to have to do a usage-based model.

I mean, like, do those factor in at all? Or this is just, it would not be possible almost to do a fixed price contract because...

I would actually argue outcomes-based is pretty different than usage-based. Okay. You know, just because think of it this way. if you have an AI agent that is making sales for Stripe to small businesses, and I told you I will sell one-tenth the number of new Stripe GMV, however you'd value that, but I'll use one-hundredth of the tokens, you probably wouldn't care. You care about the value to your top line of your business, I would argue there's not a strong correlation between token usage or utilization and value.

There may be, but there's not always. There was that infamous, I think it was called folklore, but it's this website where that Apple engineer used to put just all this Apple folklore and the folklore. Yeah, I love it. It's like if you're an engineer, it's a fun site to go to. But there was a story about some new bozo manager asking for lines of code every day. one of the engineers wrote a negative number as a way of saying like F you to the man because he refactored a code base or whatever.

I think that is the essence of why tokens are not correlated with value. They may be, but the idea that they definitely are, I don't think stands to reason. And so I think usage-based is like charging for storage or something. Outcomes-based is what business outcome is this agent designed to produce and did it produce it effectively? And that is really aligning because it creates like this whole vertical alignment. So as a company, reducing your token utilization for the same outcomes is your problem, not your customers.

And that's a great incentive to just drive more efficiencies over time. It means that to grow your relationship with the client, you actually have to make your product better. And not just theoretically better to steak dinner, you know, like better, better.

How do you have outcome-based when you move beyond customer service where there's a clear was this resolved or not to product usage where people were shopping and yeah, they didn't buy a house there, but they mostly don't buy a house on most website visits, but it was a successful visit.

So it's the right question and there's not a great way to do it for every type of agent right now. And so you can all sort of like fall back to usage-based, which is fine. But in that, over time, it's like, wouldn't it be interesting? I think AI agents should have memory. I think AI agents should drive relationships, not conversations. And it would be really interesting to say, could we make an AI agent that actually drives home ownership over time?

I think that's actually, it's hard. Yeah, yeah, yeah. But it's not.

Do you have a territory kind of?

I think so. So, I mean, even because it's hard today and, you know, we're a pragmatic company, you know, I think it's sort of the right thing to ask, though, because that's fundamentally the value the software is designed to produce. And so, you know, I think actually it's a really sort of values aligning thing. It also, though, changes the dynamics of a software company's relationship to its partners, to its clients, because if you go back ancient history four years ago, you know, There was a really stark separation between software and implementation and usage.

It was the client's accountability to use the product well. It was either your IT team or a systems integrator's responsibility to implement the software. And the job of the software company was just to make it and throw it over the wall. Obviously, it's not exactly that, but that was kind of the mark we were in. And everyone had good intentions, but what's the same success as a thousand fathers, failures, and orphan? Like when the software didn't go well, everyone was blaming everyone else.

You know, the client was like, I'm using it just fine. It was implemented poorly. The person to implement it was like, no, the platform's broken. The platform people would say, and you know, it was like everyone's pointing at everyone else. What's nice about outcomes based, whether or not the client sets it up, you become more accountable to help them be successful because until they do, they can't use it. if there is some long sort of last mile of implementation, it creates a strong incentive for the software company to have skin in the game to just help you navigate that last mile.

I think so many of the problems in the software industry are due to that lack of accountability. If you talk to any company who's ever implemented an ERP system, it's like a multi-year process. It's invading Russia. Yeah. And you don't even remember why you're doing it midway through. You've gone through two CFOs and three CIOs by the time it's done. And we're like, okay with that. That's just the way software works. And so my view is just like, I think, you know, AdWords sort of changed the advertising industry on the internet, you know, just drove it.

And I think you're going to even pay for a mobile app install, you know, now directly and truly pay for outcomes. I think it's a really positive step forward. So it's not going to be possible for everything. You have to have pragmatism, but I think it's the right way to actually have a partnership. You should share in the outcomes.

You want to wire the company to be thinking in this outcome-based way. Like in your main customer service stuff, you can do that in other ways you might not be able to yet, but you want people to be spring-loaded to be thinking that way.

That's right. If the whole company is incentivized towards outcomes, we're a way better partner to work with because of it. I mean, we find this at Stripe where, again, we have outcome-based pricing.

You've always had outcomes-based. Exactly. Yeah. Yeah, it's transactional. But we find like there's a lot of uplift we can get on just getting people more revenue and finding ways to, you know, we're sometimes hammering customers where it's like, you should be accepting, you know, local payment methods for internationalization. Or like, you're crazy not to be turning on this feature. But we really feel it because like we have the same incentive as the customer. It's like, this will be revenue maximizing for both of us.

I'm going to ask a very AGI brained question.

I'm glad we're going to second get it. Exactly.

Yeah, now we get to it, which is you described building stuff that you know you're going to throw away because the model capabilities will get there. And you're like, occasionally they are developing capabilities that you developed yourself. Isn't Sierra itself kind of shortage?

Sorry, I said I couldn't resist. No, it's the right question. You know, the short answer is I don't know. I mean, the fog of war in the software industry is pretty thick right now. I really believe in the applied AI market, though. I think most companies don't want to buy models or buy software. They want to buy solutions to their problem. And if you just go back to the cloud industry, why doesn't Amazon and Microsoft do everything for everyone?

There's not really like a sort of by somewhat similar logic, like why should any software as a service company exist when you have bigger scale, all this technology? In theory, they could just develop all the software. And actually, many of them have tried. There's actually competitors to Salesforce and almost all of the above. I think there's so much nuance in how these companies align themselves with different departments at these companies, solve their very unique problems in very specific ways.

That is a mix of product, not technology, but product, go-to-market. It's an ecosystem around it. And I think a lot of that still exists because I'm not sure coding the software was necessarily the hard part. And then similarly, I actually think, especially in enterprise software, how you engage with your clients really matters. And, you know, I think it turns out that, you know, GPT-5 and, you know, Claude, whatever version it's on right now, or Opus, excuse me, is sold to a different buyer than like the CFO or the chief customer officer or the chief digital officer.

And that seems small, but it's actually big. And so I think you tend to see software companies orient around individual buyers within companies. You tend to see consolidation around departments and around buyers. It's possible that, you know, you can go beyond those lines, but it hasn't happened traditionally. And I think the reason for it is most business users, you know, want actual solutions to their problems and they want a company that serves their unique problems in a very specific and bespoke way.

So I actually am extremely bullish on applied AI. I actually think we could accelerate. I'll make one statement, which is I think if we paused model development, we'd still have trillions of dollars of economic value. I totally agree. That have yet to be realized. And I think if we had a mature applied AI market where the CFO could go buy that agent to onboard new supply chain vendors that just worked, we could actually accelerate that trillions of dollars of economic value.

So I think not only am I somewhat skeptical that there will only be like two companies in the world, I actually think one of the main things impeding adoption of AI is the lack of existence of all those other companies. And so many of the startups, particularly around here in San Francisco, are basically doing relatively rote kind of tools around the AI, rather than actually building agents for business processes that are boring but important and valuable.

So I'm really bullish on it.

Yeah, yeah. And I guess you help companies ensure that they can always have access to the latest models, which sounds like a minor thing, but like the leading model is always changing. And so that's not a trivial thing.

I agree. and I don't know, like, I'm not sure how much of a long-term value it is. I think it is. You know, I think your customer...

Up to this point, like, the race is led by a matter of months, right?

Well, every single month there's a new frontier model and your customer experience doesn't change that frequently. So you're absolutely right. But I also think there's just a big product, right? Like, our clients use it to optimize their sales. And that is a product, not a technology. And it's very particular to the workflows of people building customer experience teams, building sales teams, and that's really what we're focused on. And I think those departments deserve purpose-built software, and I think there'll be enduring value there.

But it's interesting. It's the right question to ask. I don't think we've ever lived in a world where production of software was easy. And software engineering was the most scarce asset in a company, and now it's the most plentiful, and I don't think we've ever lived in that world.

Yes. Yes. Well, that kind of gets to one of the biggest conundrums in Silicon Valley right now is what will the shape of AI productivity be? And I think there's a strong sense that the AI has gotten really good and it should change the composition of companies and should change the hiring plans somewhat. And you're seeing this in some corners, you know, Block Doc announced their 45%, 50 % AI layoff yesterday, and you have some companies not growing as quickly.

At the same time, in coding, you see a lot of AI benefits. You can kind of argue that either way, right? You can say, engineers have gotten much more productive, therefore we should hire fewer engineers. Or you could say, engineers have gotten much more productive, the ROI on a single engineer is way higher. Like we now have super engineers that we can hire. Therefore, we should hire way more of them because there isn't like a fixed amount of stuff for Stripe or any other company to do.

And then the AI productivity story in other roles is just a bit less clear because as we've discussed, AI is kind of uniquely well-suited to coding. And so what do you make of just, how does the AI productivity show up? I feel like every company in Silicon Valley is trying to figure this out right now.

Well, first, I think I'll go back to my why I believe in applied AI. I think the atomic unit of productivity in AI is a process, not a person. I don't think AI, I don't know if you have an assistant, but if you do, he or she might help you prepare for a podcast, might help you prepare for a meeting. He or she might also get you a cup of coffee. AI will be really good at the first two, but quite poor at the last one.

So no matter of AGI short of robotics will get you a cup of coffee. So I think it's wrong to think about AI as like sort of replacing people in addition to being inhumane. It's just sort of nonsensical because AI sort of operates in the world of digital technologies. And I think if you go to like an example of even a mundane process in your business, like onboarding a new supplier, think about all the departments and people involved in that.

There's a legal department to do a contract. There's some finance department procurement to negotiate the relationship. You probably have IT that's involved to sort of onboard them into your core systems. And then there's usually a business that's sponsoring it. Fairly mundane happens all the time. Let's just say you tracked what is the median amount of time it takes to onboard a new supplier, and it was 17 days, just for argument's sake. I bet you could say as a CEO of a company I want to use AI to optimize that process and make it 17 hours or one day.

And you could go through, and if you had a product manager on that and optimize every part of it, I bet you could achieve that. But the hard part isn't like a person's job. It's actually all the systems and people in between it. And so I think part of the reason why I think it's been slow to get the productivity enhancement is we sort of ship our org charts as companies naturally. That's the natural state. There's not usually a person responsible for that process.

There's the legal team responsible for the contract. There's a procurement team. So I think actually we will end up reimagining our companies with the benefit of AI. Will we actually think of our companies as a collection of processes, have people responsible them with KPIs who can apply AI? And I think I bring it up just because that's my theory of the world. I might be wrong. I might be right. But I'm not sure our companies are set up to essentially absorb the benefits of AI efficiently right now.

And we need to do that to really do so. But the bigger point, I think, is that there's the paradox of, well, you want more software engineers. But on top of that, most of the world isn't just digital technology. And so I think a lot of the people in sort of the AGI community have only ever worked at like a research lab or a software company. you look around like wow yeah i was gonna do all of this and as they walk by the flower shop and get their coffee at the coffee shop and you think about like your the local flower shop like if you took all the ai in the world and gave it to that you gave

it super intelligence like how much would impact the flower shop's operations like maybe a little i mean i'm sure it would help yeah don't get me wrong but someone's still you know clipping the ends of the you know stems of the flowers is arranging the bouquets and, you know, thanking you on your way out the door and congratulating you for your daughter's wedding or whatever it is. And so I think if you think about, you know, what parts of the economy can absorb intelligence really efficiently, it's certainly software.

And we're seeing that already. It's finance seems particularly meaningful here because so much of finance today is just digital information. You know, we've sort of everything's in digital systems now, not even just crypto. I mean, just everything's in digital ledgers everywhere. It still doesn't touch a wet lab. It's still, you know, can't do a clinical trial. You know, you still need to get, you know, a crate from this country to that country, you know, on a ship.

So as a consequence, I think I'm not sure we'll see the productivity enhancement we see in software in every sector as quickly. And then on top of that, I think companies need to stop just giving like co-pilot to every employee and be like, we're AI now and start to think about from first principles, What are the parts of your business that have a lot of digital workflows? Where can AI have a real big impact? And how do you actually set up your company to actually have someone accountable to drive that?

And that feels like a real big change management opportunity that most companies haven't done.

Just to push on that, so software engineering, I think we clearly are seeing a lot of AI productivity gains. And software engineers have always loved tools and the latest tools and are just kind of headlong diving into it. then you have stuff like you see the flower shop where uh just and stuff that requires really good robotics that we're far away from uh that will take a while what i'm talking about is like the

there's like a by the way i might prefer a flower shop with the florist totally yeah just to say it i'm not sure it solves i'm not sure it solves the problem i have with my flower shop absolutely i might be wrong i might be unique in that yeah but i that's but i think a lot of the company

a lot of the economy is actually white collar knowledge work, not coding. Think of finance departments, legal departments, things like that, where you should be able to see a lot of AI uplift and a lot of AI productivity improvements. It just feels like a current course in speed, we're not on track to get those productivity. Well, I would, I'm not sure I'm right,

But I would argue thinking about it by department rather than by processes where it's off.

We can talk about the processes as well.

Hear me out, though, on this. Because if you said, I want to make the legal department more productive, so I want to make it easier to do red lines, and you optimize that. But why is the contract there? What is it for? You might, if you're, for example, onboarding a supply chain vendor and you have hundreds of them, you might actually say, actually making an abstract technology for your legal department to redline contracts more efficient is actually a harder, more general problem than for your supply chain vendors, because you might actually have very rigid rules around your supply chain.

Let's say you're a CPG company, and you might actually have very specific saying, like, look, if you want to work with us, here's our core legal terms, here's the axes of independence. And if you want to make an AI agent to automate that contract, that's actually a much more narrow problem domain that doesn't require general purpose redlining technology. In fact, if you sort of reduce it, you could say, well, there are like 10 % of our suppliers where we let them negotiate their contract, but only for this spend.

Let's have them go through our legal department. The rest, let's do it all with AI. And my point on it is if you look at it through the lens of like an end-to-end business process, you can turn science into engineering. And I think solving legal through AI, that's a science problem. And this is my point, though, which is I think people are going through department by department. Similarly, there's not a person accountable for that end-to-end process.

And the more you can narrow the domain that you're solving with AI, the more you can build a harness or a scaffolding with existing technology to actually fully automate it. And my hypothesis is most companies just aren't set up that way. That's just not how we're organized. And as a consequence, we're all like optimizing our site. We're all just installing Copilot. And Copilot's great, by the way. Didn't mean to insult it. But it's not actually like, yeah.

Yeah. And to be clear, like that's the kind of thing we're doing where, and obviously companies, good companies did this before AI continuous process improvement. And I feel like that is the best thing to do. And I think what you're saying is like, there's no such thing as an AI lawyer. instead of those like improving your commercial contracting. Like that is a thing that you can tend to.

And even more narrowly, like pick one domain of commercial contracting and solve that. And I actually think those are truly solvable. And I think the companies that, you know, really think about their business that way, I think they can see the value. And again, I'll go back to the immaturity of the applied AI market is probably one of the bigger barriers right now. And, you know, my hope is that as the applied AI market matures over the next few years, we'll see kind of a step change in productivity.

Yeah, there is a canonical way to build a Silicon Valley company. You have, you know, engineering and product and design. You have this number of ratios of engineers to product managers and engineering managers. And then you've got a market organization and you have these pipeline coverage ratios and you have, you know, the product marketers and all this kind of stuff. But I find it interesting how similar so many Silicon Valley companies are to each other because they've all learned from each other, right?

There's like a shared recipe and a shared playbook as to how to build a company. And it gets tweaked, but ultimately, I think it's pretty good IP. Like certainly companies are much better off with it than without. How is that canonical template for building a company different post-AI than before?

Yeah, that's a really interesting question. One is I've always believed in the primacy of tech leads over engineering managers. Both Google and Facebook, where I spent some of my early career, both did this well, where in a product review, you weren't just talking to a manager. You were talking to the tech lead and PM who were product manager who were building the product. whereas if you went to companies that produced worse software, I'd notice you sort of move up the chain of the command like the military.

Yes. I think that we will end up with individual tech leads who, because of the existence of AI agents, will become even more important, where if you are a, I'll say, product engineer, I'm trying to find the right word for it, We might invent one who has taste, but didn't necessarily know CSS, who has infrastructure ability, meaning that you understand the basics of distributed systems and debugging, and you understand your customer very deeply. With the presence of codecs, you can produce amazing results.

those people are truly worth a thousand X other people because it's relatively easy to find someone who's a great infrastructure engineer not easy easy but like relatively finding someone with good taste that's that's relatively easy find someone who also understands your customers extremely well like the nuances of the problem they're solving those people who can like combine that will, I think, end up being able to actually produce products like capital P, valuable products, with relative autonomy.

And I wonder if it will change our view on generalists broadly. I've always sort of identified myself as a generalist just because I've been both a software engineer in a suit, basically, and I've gone to kind of back and forth in that world. and as companies grow you tend towards more specialization you know just because the person who's sort of the jack or jill of all trades ends up sort of not fitting in you know like there's not really a place for them because okay well you're not really the deepest engineer you're not really the best designer you're not really a product manager if you've been at the company for a while we'll give you an honorary something to do

and you have to lead through influence and could that person actually endure as one of the most valuable people in these companies? And I think, I don't know whether it's naive optimism or true, but I actually think those people who often exist in early stage startups are often the people who get sidelined, but actually in a way that actually really harms the company. And I'm hopeful that in a world of AI agents, those generalists who, again, I think the most important part is understanding the customer need with agency, no pun intended, and empowerment can end up more powerful in the Silicon Valley company.

I've noticed the exact same thing, it's right. The exact same thing, which is high agency, really caring about customers, just really caring generally, high work ethic people who maybe weren't the best engineers previously or now, those people are massively ascendant as far as I can tell because they suddenly got the exoskeleton. And they always have the ideas as to what we should be doing and this is the better way to serve the customers and everything like that.

But now they have the way to make all their schemes real. I've really noticed that at Stripe.

Well, it's interesting you talked about work ethic. It's addictive right now because you can do so much with the technology. everyone I know who's really used it works harder because it's like wow yes I could do so much you know like you think like you're about to go to bed you're like should I get an AI agent to do something like am I wasting my the next uh you know eight hours of my life and uh I you know that might be a novelty that wears off but I think it's really exciting so I'm hopeful on the product engineering design side, you end up with these hyper high agency people who really deeply care.

I really like the way you said it actually. It's right. It's not just customer problems. It's like care, period. Just care can end up more empowered. And I'm curious what that means for organizational structures. You know, it's... I think we have a new job role we need to invent.

It's like, what role are these people in? Hyper generalists. Yeah, like kind of product managers, but sometimes maybe without a product, like minister without a portfolio, they're just doing stuff but now they can do much more.

Yeah, and it's almost like product designer, product manager, engineer. That's why I said product engineer, but that means something different. But it's interesting because we've talked about this. You know, you end up where the grass is always greener with orange structure, so you go functional organization. Okay, we're going to have engineering and product design. Let's go to business units. They're like, wow, that led to silos and war infections. We're one reorg away. Yeah, you sway back again, and, you know, that's welcome to.

Just one more reorg. Yeah, exactly. I'm a middle manager now. And I think that it is interesting if these people become extremely important, what does it mean to organize around them? And I think it does feel like something that will end up flatter just because of the amount of impact an individual can have. And so that feels really exciting to me. But it sort of feels like a blurry picture right now. Enhance. I agree. It's very blurry.

It's so interesting.

You were on the Twitter board during the super interesting takeover battle with Elon Musk. What are your reflections on that experience a few years later?

It was really interesting to sort of be in the public spotlight. I hadn't really experienced that in my career before. I joke like no one really cares about enterprise software. I worked for Salesforce for six and a half years. Sorry, what's the joke? And I worked for Salesforce for six and a half years, and I don't think my mom knows what Salesforce does. And so to have something that was not really just like a business issue or a technology issue, but like sort of in the mainstream, I realized I didn't love that very much.

You prefer enterprise software? Exactly. I'm like a builder. I like to build things and have people use them. I know it sounds sort of funny and reductive. That's what gives me joy. So the one thing I realized is the conflict of it all, however it turned out, like victory, defeat, whatever it was, it just wasn't something that filled my bucket very much.

What do you make of the fact that in all these kind of headcount debates, Elon is now running Twitter with 80, 85 % fewer people? I think Nikita Beard tweeted recently that all of Eng Product and Design at Twitter is 50 people. And, you know, maybe it's been a little flaky in pockets or just at times, but mostly the service works. And they have shipped new features. And I think those two statements are undeniable. But just what's your takeaway from that?

I don't know. I haven't followed as much as sort of the, like, I didn't see that tweet as an example. Do you call it tweet still? Sorry, I'm a little fashion. So I don't know about that, but I mean, it is interesting right now because obviously a lot of that predated AI. But I mean, any person who's been an individual contributor engineer knows that the size of the team does not produce like linearly greater outcomes. Everyone in the world has experienced that.

So, you know, I think the, you know, the idea of can you actually give individuals with good taste more agency, no pun intended, I think it's always been sort of an enduring thing. What was Jeff Bezos, a two pizza box sort of thing?

But then do large tech companies underrate this phenomenon? Like, do they pay lip service to small empowered teams and two pizza teams plus extras that maybe they should be doing more?

I think their companies largely act somewhat rationally. I can't remember who the CEO was, but it might have been the Rippling CEO just talking about, you know, there's this idea of being like lean and agile. And then there's like, you want to capture market share and, you know, grow your product and grow your platform. And, you know, at the end of the day, you can be clever, but not smart. And, you know, you might be so clever to think like, I'm not going to have anything more than two people on these features.

And if you have a competitor who maybe does something a little less elegantly, but wins, like who cares that you are clever with your, you know, two pizza box team or two person team or, you know, one AI agent team or whatever it is. when someone said we're going to have like a x billion dollar company with one person i think that might have been right but it's not you could have had 10 billion dollar company if you'd hired a bit more that's right and i would actually argue the more specific thing is if all of a sudden for some uh you know clever reason you want to prove you can the idea that like a competitor might have

10 people and beat you is probably more likely than even having 10 billion dollar company And so I think at the end of the day, when you're building a business, especially one that's in hyper growth, which successful businesses in tech tend to be, if you are too clever and austere, and going back to your point about Silicon Valley cultures all being the same, there are examples of companies that really innovated in culture. You know, you wouldn't think of it this way, but like HP, sort of like a lot of the kind of traditional open office floor plan, you know, came from them.

Facebook.

Laws worked as HP.

Oh, I didn't know that. That's interesting. And then, you know, Google offered free food to their employees, which a lot of people did. And then, you know, Facebook, you know, a lot of like the layouts of offices all looked like Facebook for a long time. But then you have other companies like we're going to innovate in HR. And they spend all this time and energy on it. And in fact, the smart thing to do is just be like, we're not, it's not what we do.

Let's just do the same old thing as everyone else because everything is just, it's push button. I don't need to worry about it. And so I think, I do think it's the right question to ask for every technology company.

Yeah. After being on the Twitter board during the Elon takeover, you were then on the OpenAI board when Sam got fired. Have you considered that you are the problem? You are bringing the drama. I came in after the drama there. Oh, right, you joined after. Oh, sorry. I was brought in as the mediator.

I see.

Yeah. Post the, okay, yeah, okay, your hands are clean.

I'm aligning my reputation here. Yeah, I was, I wasn't actually on the other side of it, but I was, I got a phone call, was it Saturday or Friday after? And, you know, basically my understanding was I was the person that both the existing board and Sam agreed upon to kind of help mediate the situation.

What have you learned in the OpenAI board?

A lot. I mean, certainly the most interesting part is the AI research. I've never been affiliated with a true research lab before, and that's fascinating to me. It is very inspiring. I mean, it's very easy to grow, not cynical, but like, you know, you can look at, you know, OpenAI, Google Anthropic and say like, who's, you know, whose model scores better on this leaderboard to actually go in and see this company where every single researcher trying to make safe AGI and not come out of this board means Inspired is impossible.

Like, it's amazing. The other thing is it's the first not-for-profit board I've been affiliated with. um and uh that's really interesting as well just because different thing yeah well i mentioned the fiduciary duty is you have a duty to the mission yeah and uh that is really clarifying and interesting as well because when you're making decisions and you realize you know you have your sole duty is to ensure that artificial general intelligence benefits humanity um that's really different it's really interesting i've never had a fiduciary duty to a mission before so that's really interesting to me because I take those duties really seriously and like reflecting in a board meeting and you're making a decision, you think about

it very differently through that context. And then the other thing was because I was brought in, you know, after that crisis, there was three people on the board on the other side of that when I agreed to temporarily be the chairman, it's still there. We had to grow the board essentially from scratch. And so that was really interesting too, just to think about, normally you add one board member at a time. This one was like, do you have a bulk rate?

We're going to build a board, put together a team.

So you really think about, spend time with the other two board members, just really thinking about what is the composition for an open AI board look like? How do you represent the not-for-profit part of it? How do you represent safety? How do you represent the economic impact of AI? We're doing lots of infrastructure investments. How do we find someone with that specific type of financial expertise? And so that was really rewarding as well, just sort of building a board, not from scratch, but effectively from scratch.

Last question. What are your AI predictions for 2026?

I think we will have some scientific breakthroughs with AI that positively break through into the mainstream press and awareness. We've already had some interesting math proofs, but I joked with one of my friends, like, until I can understand what the title means, I'm not sure it's going to make the—

You're not excited about an dimensional manifold space.

Exactly. And, you know, it won't quite be like the, you know, Apollo landing. But, you know, I remember, you know, the Kasparov, you know, chess match. And I certainly, you know, things like AlphaGo were really meaningful. I, given the progress in math, I'm hopeful we have at least one moment of discovery that is inspiring. Because I think a lot of the dialogue around AI right now is economic opportunities, but also what could go wrong. And I actually think one of the main things that can go right is actually discovery in science that actually can improve the human condition.

So I'm really excited for it because I think it will contextualize why so many of us are excited about this technology in a way that sort of captures attention. So as you said, something beyond n-dimensional manifold, blah, blah, blah. And I feel not confident in that, but it certainly feels like the ingredients are there for that. I think we'll continue to see mainstream adoption of AI by both consumers and companies. that doesn't really feel like a prediction.

But I think this will be really a year of adoption of agents. And we're certainly seeing that in Sierra's customer base, but I think we're going to see it more writ large. And then you're already seeing chat to PT growth, you know, really unprecedented levels. And things like OpenClaw, you can sort of see that kind of translate over to agents and sort of the more like long-running autonomous tasks. So it does feel like by the time we exit this year, can that go from a niche community to something more mainstream?

It feels probable to me. And then the other thing is I think most companies in Silicon Valley won't write code by hand. And that might seem almost, it's sort of funny that it sort of feels obvious right now. You're like, oh, yeah, of course. Like you're just nodding. Like, yeah, of course. Yeah, why not? But if I had said that like four months ago, that would have been a bold prediction. Yeah. But I think that's really interesting just because that's such a fundamental state change.

And I say in Silicon Valley because I do think it takes a while for these tools to diffuse through society. Silicon Valley is insular enough that I think it will here, but I'm not sure it will happen through every company in the world yet.

So the year of agents across businesses and just people finally getting their kind of clothed agents and then yeah, all clothed written by AI as well. Yeah. It's good celebrations. Thank you. Thanks for having me.