11 Sep 2025 · 55m · Bret Taylor, Alex Heath · 23 quoted facts
Bret Taylor, CEO of the AI customer-service startup Sierra and chairman of OpenAI, argues that AI is both a real economic transformation and a bubble, much like the dotcom boom. Some companies will lose huge sums, while others become giants. He says AI agents can do real work for large companies, and that charging only for successful outcomes is the right way to sell them.
The central tension is between huge promise and real uncertainty. Taylor sees phone calls, coding and customer support being transformed, with models already reaching what he would once have called AGI. Yet he admits nobody knows what happens to the software market, and he expects investor losses. He rejects the MIT study on failing AI projects, arguing that buying finished AI products works better than building your own.)
Sierra charges a fee only when its AI agent resolves a case on its own, and transfers to a human are free. Taylor says this forces a closer partnership with customers than traditional software sales, where the vendor is paid up front.
Taylor says a human-answered call costs roughly $10 to $20, more than many brands earn per customer, so companies discourage calls. AI cuts that cost by about two orders of magnitude, so brands can talk with far more customers and may win market share.
Taylor says voice became a bigger part of Sierra's platform than text within a year of launch. The phone is one of the oldest analog channels, and AI is making it digital for the first time.
He argues that systems pass the Turing test, which asks whether an AI is indistinguishable from a human in conversation, and that the goalposts keep moving. He sees the remaining gap in physical-world tasks and scientific breakthroughs.
Taylor compares AI to the internet boom, where Pets.com and Webvan failed while Amazon and Google thrived. Both outcomes can be true at once, and overbuilt infrastructure like fiber was later reused by others.
Sierra's customers pay a fee only when its AI agent fully resolves a customer's issue on its own. If the agent has to hand the customer to a human, the company pays nothing, according to Bret Taylor.
Taylor says a phone call costs roughly $10 to $20 in labor and technology, which is often more than the company earns from that customer. AI cuts the cost sharply, which is why he thinks companies can now afford real conversations.
Yes. He says it is a bubble in which many people will lose a lot of money, much like the dotcom era, but he also thinks AI will create huge economic value, as the internet did.
He says the study lumped together companies building their own AI software with companies buying finished products. He argues that buying ready-made solutions, like Sierra or the legal AI firm Harvey, succeeds far more often.
AGI means artificial general intelligence, AI that matches or beats humans broadly. Taylor says current models already meet the definition he would have used three years ago, at least for digital tasks and ideas, though he expects scientific breakthroughs still to come.
Stated by the speakers. Each quote matches the transcript, which shows what was said, not that it is true.
Bret Taylor was an early Google engineer who founded the social network Friendfeed, sold it to Facebook in 2009, and then served as Facebook's CTO.
“He was an early engineer at Google before starting Friendfeed, a social network he sold to Facebook in 2009, where he then served as CTO.”
Bret Taylor led Twitter's board during Elon Musk's takeover and became chairman of OpenAI's board after Sam Altman was fired and rehired.
“he led Twitter's board during Elon Musk's takeover and became chairman of the OpenAI board after the firing and rehiring of Sam Altman”
Sierra raised a new funding round in the week of the episode that valued the company at $10 billion.
“Just this week, Sierra raised a new round of funding, valuing it at $10 billion.”
Bret Taylor says he announced he was leaving Salesforce within days of ChatGPT's release.
“I happened to announce I was leaving Salesforce within a few days of ChatGPT coming out.”
Bret Taylor says he decided to start Sierra during a lunch with Clay Bavor, whom he had known for twenty years, without having planned to found a company together.
“ended up having, uh, lunch with Clay Bavor, who I'd known for twenty years and not planning to start a company with him”
Bret Taylor says Sierra's customers include ADT Home Security, Ramp and Sirius XM, which use its AI agents to answer customer calls.
“everyone from ADT Home Security to Ramp in New York to Sirius XM are using it to answer the phone when you call them up”
Sierra charges most customers a fee only when its AI agent autonomously resolves a case, according to Bret Taylor.
“we charge only for outcomes. So for most of our customers, that means when the AI agent autonomously resolves the case that you called about or chatted in about, there's a fee for that.”
Bret Taylor says Sierra does not charge when its AI agent has to transfer a conversation to a human.
“And if the AI agent has to transfer to a real person, it's free.”
Bret Taylor estimates that answering a customer phone call costs roughly ten to twenty dollars in labor and technology.
“say it's somewhere between ten and twenty dollars to answer the phone”
Bret Taylor says large language models reduce the cost of a phone call by roughly two orders of magnitude.
“you bring down the cost of a phone call by not one, but two orders of magnitude”
Bret Taylor says some Sierra retail customers let shoppers photograph damaged goods and have an AI agent approve a warranty claim immediately.
“We have retailers where you can take a photo of your damaged goods and immediately adjudicate a warranty claim”
Bret Taylor says voice is already a bigger part of Sierra's platform than text, although voice launched only the previous November.
“voice is already a bigger part of our platform than text, uh, which is kind of remarkable since we launched it in November of last year.”
Bret Taylor says Sierra fine-tunes existing AI models but does not pre-train its own.
“We do a lot of fine-tuning. We don't pre-train any models.”
Bret Taylor says a single response from a Sierra agent typically requires twenty or more inference calls to AI models.
“For any given message for to one of the agents on our platform, probably twenty plus inference calls just to generate one response.”
Bret Taylor says current AI systems already meet the definition of artificial general intelligence he would have used three years earlier.
“I think we are already at what I would have defined AGI at three years ago.”
Bret Taylor defines superintelligence as AI that is more intelligent than humans, and says the term matters for safety research.
“Superintelligence, I think, literally just means that it is more intelligent than humans.”
Alex Heath quotes Sam Altman as saying at a dinner that someone will lose a phenomenal amount of money in an AI bubble while many others will make a phenomenal amount.
“Someone is going to lose a phenomenal amount of money, we don't know who, and a lot of people are going to make a phenomenal amount of money.”
Bret Taylor cites Pets.com and Webvan as famous failures of the dotcom bubble while arguing the internet still produced giants like Amazon and Google.
“a lot of people think about the flops like pets dot com and WebVan.”
Bret Taylor notes that many companies that built out fiber in the early internet era went bankrupt, but the fiber was later used by others.
“the people who built out fiber in the, um, early days of the internet, a lot of them went bankrupt, but that fiber ended up getting used”
Bret Taylor says a story about him rewriting Google Maps over a weekend is mostly true and a little embellished.
“there's this like story that goes on around me about rewriting Google Maps, and and it's like, mostly true and a little embellished, like many great stories are.”
Bret Taylor cites the film Hidden Figures, about women who did the math calculations for the moon landing, as a parallel to software engineers facing automation.
“Hidden Figures. It's a great Disney movie about the, uh, putting people on the moon, but in particular, the women who did the math calculations to do it.”
Bret Taylor criticizes the MIT study on enterprise AI spending, saying it conflated companies building their own AI software with companies buying ready-made AI products.
“That study, I don't know the basis for the, um, data in that study. I think it was problematic because it conflated people building their own software with AI and buying solutions off the shelf.”
Bret Taylor says Sierra has a roughly 100% success rate moving customers from proof of concept to going live.
“We have, uh, basically 100% success rate with our customers doing a proof of concept and going live with our platform.”
Taylor explains how ChatGPT's arrival and a lunch with Clay Bavor led to founding Sierra. It shows how a technology shift can reopen entire markets.
“I happened to announce I was leaving Salesforce within a few days of ChatGPT coming out.”
Taylor explains why Sierra charges only when an agent resolves a case, and how that changes the vendor-customer relationship.
“we charge only for outcomes”
He lays out the economics of call centers and how cheaper conversations could let big consumer brands talk to customers they could never afford to serve.
“Big companies have big problems.”
Taylor gives examples such as warranty claims and refinancing, and argues that guardrails and regulation, not basic question answering, are the hard part.
“Slapping ChatGPT together with a knowledge base is not that hard.”
Voice is already bigger than text on Sierra's platform. Taylor argues AI is turning the phone into a digital channel.
“voice is already a bigger part of our platform than text”
Sierra fine-tunes but does not pre-train models, and Taylor compares model choice to picking a database.
“We do a lot of fine-tuning. We don't pre-train any models.”
Taylor says today's systems already meet earlier AGI definitions and explains why the field now talks about superintelligence and safety.
“I think we are already at what I would have defined AGI at three years ago.”
He agrees AI is a bubble but argues it will still create huge value, as the internet did despite the Pets.com and Webvan failures.
“And I think we're also in a bubble, and a lot of people lose a lot of money.”
Taylor rejects the MIT study and says companies will buy specialized AI agents rather than build their own.
“We're just in a really new world because we're taking something scarce and making it plentiful.”
As soon as autumn rolls around and you swap your iced lattes for pumpkin spice, that's your signal to start thinking about the holidays, the decor, the gifts, the shopping lists, but more importantly, the holiday treats, cheesecakes, cookies. Is there anything better? And when the recipe calls for cream cheese, there's only one option Philadelphia, the original cream cheese, made with fresh milk and real cream. This holiday season, whether you're going all out or keeping it simple, make it really Philly good.
Find your favorite holiday recipe at creamcheese dot com.
Running a business shouldn't feel like surviving a software group project. One app for accounting, another for inventory, another for sales, and somehow none of them talk to each other. That's where Odoo comes in, an all-in-one business management software that brings every part of your business together. From sales and accounting to inventory and marketing, all in one powerful platform. No messy integrations, no bouncing between tabs, and best of all, no spreadsheets. Stop managing software and start managing your business with one unified system.
Try for free today at odoo dot com slash vox. That's o d o o dot com slash vox.
Support for the show comes from CrowdStrike. It's not a huge stretch to say that AI is the next major computing platform. Every platform shift changes the way we build software, and it changes security, too. CrowdStrike is defining cybersecurity in the AI era with AI detection and response, or AIDR, a solution for seeing, monitoring, and securing AI across the enterprise. Learn more about how leading companies are turning to CrowdStrike to secure AI and secure their business at CrowdStrike dot com slash decoder.
Hello and welcome to Decoder. This is Alex Heath. For my final episode as your Thursday episode guest host, I recently sat down with Bret Taylor, the CEO of AI startup Sierra and the chairman of OpenAI, for a live event in San Francisco hosted by Alix Partners. Very few people have seen the tech industry up close like Bret Taylor has. He was an early engineer at Google before starting Friendfeed, a social network he sold to Facebook in 2009, where he then served as CTO.
He then founded Quip, which he sold to Salesforce. After eventually becoming co-CEO of Salesforce, he left to start Sierra, which is rethinking how businesses use AI for customer support. Along the way, he led Twitter's board during Elon Musk's takeover and became chairman of the OpenAI board after the firing and rehiring of Sam Altman. As you'll hear in our conversation, Bret is all in on AI. Just this week, Sierra raised a new round of funding, valuing it at $10 billion.
In this episode, we get into the origins of Sierra and what it's doing now with AI agents. I also asked Bret about OpenAI and the overall trajectory of the AI industry. We covered a lot of ground, and I hope you find Bret's perspective as fascinating as I did. Okay, here's my conversation with Bret Taylor. Bret, thanks for being on the show.
Thank you for having me.
I would like to start by going back to early 2023. You're leaving Salesforce, you were the co-CEO. Talk about that process of deciding to make a new company, what you looked at, why did you land on Sierra at that time?
I happened to announce I was leaving Salesforce within a few days of ChatGPT coming out. So I don't know if there's like, you believe in like cosmic forces, but I was like every single human being, particularly a geek like me. When you first use a product like that, it's all you can think about. So I was honestly not 100% sure what I wanted to do when I left, you know, I was trying to leave and then figure it out, which is a good good thing to do in life.
And I immediately just became obsessed with the technology. I was using it personally. Reed Uh, Reed Hoffman's a friend of mine and was showing me early versions of GPT four, and I just couldn't believe the level of empathy, and it just truly sounded human. And I had been following AI for years, but I honestly, you know, uh, if you had told me in whatever October, the month before, to define what a large language model is, I would have given you a blank stare, you know, and ended up really realizing that, uh, this technology, which I had not been following as closely as I I wish I had, I really came to believe was gonna change, um, change the world.
And so I knew I wanted to work in it, I didn't know what I wanted to do, but it that was okay. It reminded me a little bit of when I first discovered the internet, you know, I think you, you know, it's gonna change everything. At least I felt that way, and, uh, I was excited to work in that space, and that's all I knew. So I ended up having, uh, lunch with Clay Bavor, who I'd known for twenty years and not planning to start a company with him, but found out through the course of the lunch that he was equally obsessed.
Who was working for Sundar at Google at the time, and by the end of lunch, we had a couple more courses than we had originally planned. Uh, we had decided to start a company, and we had no idea what we were gonna do. But I think it was really based on the premise that when you have a seismic shift in technology, a lot of business opportunities present themselves because it kind of shuffles the deck of both what consumers want, what companies need, and what software vendors are adequate to support that that need.
Um, you know, if you look at the advent of the internet, it gave birth to some of the largest names in in the stock market today, like Amazon and Google. It disrupted companies like Microsoft, who got through it quite strong, it disrupted companies like Siegel systems, a little bit less strong, right? And so you end up where the incumbent insurgent dynamic changes quite a bit, and huge markets open up. In the case of the internet, it was search and e-commerce, and I think with large language models, and we we'll probably talk about it more, so I won't, uh, have too long-winded of an answer here.
Uh, it really stands to reason that, you know, a lot of different markets, from software engineering to customer service, are going to be transformed, completely upended. What an interesting time to start a company. So we left and gave ourselves a, you know, a few months of just recovering from our jobs. And then talked to a lot of customers and decided to, uh, build Sierra, and at Sierra, we're building AI agents for customer experiences. So, uh, everyone from ADT Home Security to Ramp in New York to Sirius XM are using it to answer the phone when you call them up, or in their digital properties, uh, have a conversation.
Doing everything from helping you upgrade or downgrade your Sirius XM plan to calling you when your ADT alarm goes off, which I think is, uh, pretty, pretty exciting.
Yeah, talk to me about Sierra and how you work practically with a with a new customer. Like, walk me through that process, cause this is all a very new field. I mean, customers customer support is not new, but the way you're doing it is new. So what is unique about how you work with a customer versus, like, how you would have done it at a at another company before?
I'll start with our business model because I think it helps answer your question, which is, um, one of the things that we do differently than traditional software companies at Sierra is we charge only for outcomes. So for most of our customers, that means when the AI agent autonomously resolves the case that you called about or chatted in about, there's a fee for that. And if the AI agent has to transfer to a real person, it's free.
Uh, we really like this as a business model, and I think it will become the standard business model for agents because, uh, the word agent comes from agency, and the principle of it is some degree of autonomy. And, uh, I think most of the most sophisticated agents will actually start and complete a task, whether it's generating a new lead for your sales team or solving a customer service inquiry, or, uh, doing a legal analysis, uh, for an antitrust review, whatever it might be.
And if an AI agent isn't just helping a person be more productive, but actually accomplishing a task, why not just pay for a job well done? Uh, and if you look at most of your companies, the if you have a job where the outcome is measurable, like sales, you tend to pay a commission, right? There's not, uh, just a salary. And so I think agents sort of being paid on commission, if you will, is actually not only a great incentive alignment between a a vendor and a partner and a and a company, but also just feels, uh, from first principles, right?
And that's why I think just like the advent of cloud-based software, when Mark and Parker started Salesforce and it was a subscription-based service rather than a perpetual license, changed the landscape of software. I think the same will happen with agents. Now, going back to how we work with customers, it kind of ch- uh, begs the question of what your relationship is between a software vendor and a company. If you only get paid when it works, there's a certain degree of, like, arm's length relationship that most software vendors have with their customers.
And if you've, you know, ever seen someone who's done a big ERP implementation, I don't, I don't know much about ERP systems, but apparently it's really hard to implement because everyone I've ever met who's done it, it's taken like two years longer than you expect and a lot more money than you expect. And if you go and talk to the 10,000 people involved in one of those projects, the systems integrators pointing to the software vendor, the software vendors pointing to the systems integrator, no one's really pointing at the company because they're the one paying the bills.
So everyone's like, Oh, you're great. No, every I'm sure everything's fine. And you know, it's like, you know, success has a thousand fathers, failures, an orphan. But part of the issue is, uh, the only party in that relationship that cares about the outcome is the company. And so everyone's blaming everyone else, and, you know, it requires a good CIO or CTO to navigate that. But you can see all the perverse incentives involved, you know, maybe the partner's getting paid by the hour.
Not great, not a great incentive. The software vendor's already made the sale, so, like, good luck to you getting it successfully deployed. So I think going towards outcomes-based pricing demands a different relationship between a software company and the companies that they work with. Uh, I heard, uh, you just talked about forward deployed peeps, and, uh, yeah, I'm not sure that is the only answer. Um, I think it's trendy right now in AI, in part for this reason.
No software company wants to be a professional services firm, so it's you can't turn that knob all the way up to eleven and build the company that I think you want to build. But there is a different level of accountability. And so, you know, with our relationship with our customers, we've really focused on sort of, uh, a couple different things. One is, uh, product ease of use. Um, I think to make your outcome, you need to make it as easy as possible to achieve that outcome.
So we somewhat uniquely in the market have a product for technology teams as well as a product for operations teams. So you can build agents without any technical knowledge at all, and again, trying to empower as many customer experience professionals as possible. And then on the partnership side, we have a lot of support with what we call agent development. So if you need help getting your agent out the door, we show up in a bus to help you do it.
And that's really unique. Um, what I really, I I'm not sure how everything will play out, but I am I'm really bought into this vision. And when I talk to our our customers, I love the idea that, you know, they know exactly the value that we're providing for them because they only pay us when the agent works. And I just love the simplicity of that relationship, and I'm I'm really bought into that.
So you have hundreds of customers, and 50% have revenue over 1,000,000,020% over 10 billion a year. Is that right?
That's correct.
Why focus on big customers for like that, instead of, you know, a huge Shopify-like approach to this? Why are you going towards the the the big companies?
Big companies have big problems. I love first principles thinking, and if you are a large consumer brand and you have 100 million consumers around the globe, before large language models, you literally could not have a conversation with all of them. Um, if you just do the math, uh, there's a term in call centers called cost per contact, and it's essentially measuring how much all in labor costs and technology costs to answer the phone or answer the chat.
It really depends on how complex the conversation is, what the qualifications of the person answering the phone, it depends whether it's onshore or offshore. But say it's somewhere between ten and twenty dollars to answer the phone. For most consumer brands, their average revenue per user is less than that phone call, uh, and so you literally can't afford to have a conversation. It's why if you've ever tried to call any consumer brand, you can't, you know, there's entire websites devoted to finding the phone numbers for many consumer brands.
And it's not because they don't care about you, it's like they just can't. It's just not economical. If they try if you, if everyone who wanted to call them called them, they would go out of business, which is probably not good for you either. Now, with large language models, that's totally different. Uh, you bring down the cost of a phone call by not one, but two orders of magnitude, and all of a sudden, the economics of having a conversation change dramatically.
And so the reason why we've pursued larger enterprise brands is that's the type of, I would say, step change function in customer experience that is relevant to a company that has tens of millions or hundreds of millions of customers, and and those are the larger, uh, companies in the world. And what's really exciting about it is I think a lot of people think when they think about AI agents for customer experience, they think contact center automation, and that's a huge part of it.
But if you think about it through the lens of what I just said, you now can have an order or two orders of magnitude more conversations with your customers than you could before for the same cost, and that's really remarkable. And if you think about all the companies who are competing for, you know, whether it's a let's say a mobile phone company, you're competing for sort of a fixed pie of of customers trying to decide which which company you're gonna align yourself with.
And if you can improve customer retention by a hundred basis points, that's a lot of value. Um, if you can reduce your attrition and churn by, you know, 500 basis points, that changes the lifetime value equation of your company. And so I think people are thinking about it. I think, really, the first order effect of reducing the cost of, say, a phone call, which is great. You can, you know, uh, save that money, return it to shareholders.
But I think the more sophisticated companies are saying, Can I actually gain market share? And that's really, really exciting, and that's what we're trying to do for some of the largest brands in the world.
We need to take a quick break. We'll be right back. Support for the show comes from Vanta. As AI adoption grows, so do your company's security risks and requirements. Most compliance tools promise automation, but you're really stuck doing all the work by hand, chasing screenshots, validating controls, and playing catch-up when audit season comes around again. Vanta works differently. Their agentic trust platform is built to scale with you, not slow you down. With over 1,400 automated tests across more than 400 integrations, Vanta collects evidence and monitors your controls year round.
Plus, you can access the Vanta agent everywhere you work, whether your team is in Vanta every day or living in tools like Claude or Cursor. They work around the clock from full program context, so nothing slips through the cracks. The numbers back it up. Vanta delivers 526% ROI over three years and pays for itself in three months. Over 16,000 fast-growing companies like Ramp, Writer, and Harvey save time and resources while scaling fast with Vanta.
Learn more at Vanta dot com slash decoder, spelled Vanta dot com slash decoder. Support for the show comes from Rippling. You just found out your sales team is at risk of missing quota. Don't panic. Just ask Rippling AI. Since it's built on your real-time people and business data, Rippling AI can pull metrics from Rippling and Salesforce into a meeting-ready dashboard, which shows quota attainment, headcount plan, and monthly revenue to quota by region. In seconds, you'll see exactly what's behind your quota risk and fix it before it's missed.
Question answered, action taken, crisis averted. When you have critical business questions that need answers, don't just file a ticket and wait weeks for an outdated report. Describe what you need and have Rippling AI build it instantly from your live people and business data, whether it's a dashboard with detailed charts or automated workflows with the right triggers, conditions, and approvals. Ready to rule your business? Head to Rippling dot ai slash decoder. Get the only AI built to give you full visibility and take complex actions across your entire organization.
That's r i p p l i n g dot ai slash decoder. Sign up for exclusive access today. Support for the show comes from Granola. Have you ever left a meeting thinking that was productive, and then ten minutes later, when it's time to take next steps, you're at a loss? Or worse, you've been heads down the entire meeting, furiously typing notes instead of actually listening, and now your brain is fried. Meetings are a mess. Granola fixes that.
Granola is an AI-powered notepad that makes your meeting notes useful. It capture what happens in your meetings and turns it into clean, structured notes with decisions and action items pulled out and made easy to find. And the best part, Granola works through your device's audio, which means it integrates seamlessly into the video conferencing tools you already use. No setup, no awkward bots, it's just your normal meeting with superpowers. You get to actually listen instead of frantically typing every word, and still walk away knowing exactly what was decided, who's doing what, and what comes next.
Once you try it on your first meeting, it's hard to go without. You could try it totally free. Just head to granola dot ai slash decoder. That's granola dot ai slash decoder to get your time back.
We're back. Do you have agents right now that are doing things for customers without human involvement? I'm talking like beyond a a chatbot, but actually doing things that have, uh, economics tied to it. Or that would be something that you would think a human would be involved in, but it's actually not. Is there an example of this right now?
I'll give a few. Uh, we have retailers where you can take a photo of your damaged goods and immediately adjudicate a warranty claim, and it will connect to the inventory system and ship you a new product. We you can refinance your home with an, uh, AI agent powered on our platform, uh, end to end, um, without
Without a human in the loop?
without a human in the loop. The these agents are remarkable, um, at what they're doing, and not only are the, uh, can you take action with an agent built on the Sierra platform? Literally 100% of our customers are doing it. To some degree, there's this technique in AI called retrieval augmented generation, which is a fancy way of saying answering questions. It turns out that that's kind of a commodity at this point. Slapping ChatGPT together with a knowledge base is not that hard.
You know, most engineers nowadays could literally do that in a weekend, which, by the way, is mind-blowing. It was science fiction three years ago, now it's a weekend project. That's welcome to technology. It's mind-blowing. Actually being able to put sophisticated guardrails around a regulated process. Uh, we work in health insurance payer industry, uh, we work in, uh, the provider space, uh, we work with banks, we work with, uh, property and casualty insurance companies. And if you're talking about, you know, sophisticated regulated conversations like claims processing, that's not retrieval augmented generation, right?
That's a a very complex conversation with regulatory oversight. How do you put, uh, AI-based guardrails around it? How do you put deterministic-based guardrails around it? How do you solve the mundane problems of transcription accuracy and forty plus languages? Um, and you know, it turns out transcription accuracy doesn't really matter if it misses the word and or or, but it really matters if it's your account number. And so, you know, how do you get the, uh, hard parts?
We do roadside assistance, uh, and it turns out, uh, if you've ever chatted with an AI agent and a car horn honks, it'll often stop talking, cause it thinks it's being interrupted, cause it can't disting- distinguish the difference between a car horn and you talking. And, you know, our platform is really designed to solve those problems. Effective guardrails, multilingual conversations over chat and voice, deterministic guardrails, uh, AI-based guardrails, which is called supervisor models, which is really, really effective and interesting, and simple stuff like knowing, Hey, that's the television on in the background, no one's talking right now, or that's a car horn, someone's not interrupting me.
And it turns out that, you know, I'm sure in three or four years, that'll be easy. Right now, it's really hard, and so, which is why why we have a lot of demand for our product.
I'm glad you brought up voice. I'd be curious to hear how voice is entering this mix beyond chat and where are you. Do you think voice will actually be a bigger piece of the pie for agents than than text?
I do. Uh, voice is already a bigger part of our platform than text, uh, which is kind of remarkable since we launched it in November of last year. Uh, I think it stems from a couple of reasons. One is if, you know, first, I'll just go to, like, human parts of it. I mean, if you watch movies about computers in the future, um, or or vision science fiction authors' vision of the future, you're usually talking to a computer.
Um, I think it is the most ergonomic interface we're we're all born with it, we all know how to talk, and as a consequence, I think, uh, it is quite low friction, it is quite accessible. We talk a lot about the digital divide, and I think if most of the ways you interact with digital technology is just speaking, what a great way to just make it accessible to everyone, especially if it's multilingual and patient. And then if you look at, you know, the telecommunications industry, the health insurance industry, you know, things like that, a lot of it still goes over the phone.
Um, and you know, it's not just, uh, as patients or consumers, but, you know, payers to providers, uh, providers to payers, excuse me. You know, a lot of this is still running over the phone. And what AI has done is it's taken one of the oldest analog channels, which is the publicly switched telephone network, and made it digital for the first time. Uh, you know, it used to be that if I almost every company I talk to has a digital self-service team, which is a fancy way of saying, Hey, can you please use our website rather than calling us?
Cause, gosh, it'd be better for you, it's better for us, it's cheaper, it's faster, and, like, there's entire teams devoted to that. Now it's like, maybe call us, it's all good. It's the same. It turns out the same agent on our website is is picking up the phone, which is kind of crazy. So you've essentially, you know, you always talk about like, uh, you know, TCP IP, which is like TCP running over the internet protocol. Like, there's some name for this.
We've basically put the internet on the phone, uh, like that, we've just made the phone a channel for digital technology for the first time. And so as a consequence, you know, if you look at there's a proverb in in, uh, entrepreneurialism, like, you wanna make a painkiller, not a vitamin, because people buy painkillers and people think about buying vitamins. This is truly a painkiller. Like, you know, you've just taken the most expensive, the most tedious channel, and it everyone hates it too, by the way.
It's not even if you talk to the best customer service agent of all time on the phone. It's usually after you've been waiting on hold for ten minutes, and because the economics of not making a a call center where you don't have to wait on hold are just untenable. So it's just one of those things where, like consumers, companies like, there's no one defending the current landscape of phone calls at all. Everyone hates it on all sides, and now you have this technology that just solves the problem.
And so that's why I think it's gonna have a big impact. But looking forward, it's really unclear. I don't think most people, and I'm kind of in the center of a lot of this AI stuff, I couldn't tell you where the world is going, but I I think it is really exciting. If you look at the way, you know, WhatsApp is used in Brazil and India, you wonder with conversational agents, you know, will that style of of digital interaction be as pervasive in other markets?
Um, I was blown away when I went to Brazil, I don't know, four years ago or something, and saw someone do a mortgage over WhatsApp, and I was like, tell me what you're doing. It's like uploading a PDF. And all of a sudden, you know, if you think about every company in the world's customer experience having a conversational agent, yeah, maybe every company in the world will have a WhatsApp presence to do that, or, you know, maybe smart speakers will make a a comeback.
I think about driving into work and CarPlay and and CarPlay. I love the product in a lot of ways, but you can't really do anything with it. Imagine triaging your email, having a conversation with a personal agent while you're driving into work, and all of a sudden your commute just got super productive. Like you'd be talking to, like, the the, uh, the personal assistant with a PhD. And everything. I mean, that's pretty cool. So I think what's exciting, we talk about phone, because I do think it is the area that is just economically impactful right now.
We're making computers conversational, and I think it is a user interface paradigm as much as a technical, uh, uh, change. And, um, actually, I've never felt so old. The other day, I was talking to the phone, uh, like this for people online. I'm touching the phone to the side of my face, which until this moment I thought was normal. My kid was like, you're touching your phone to your face. It would
Yeah, that's right.
be like someone licking their phone or something. And he like, all kids just talk on the phone differently. And I, I never thought of that as abnormal until that moment, and then I was like, I'm fucking old, um, and excuse me, I don't I'm
No, it's fine.
allowed to say this in
It's fine.
the podcast. And you realize that I just think that, like kids who grew up with these technologies, who never see a rotary dial, they just have a different, like, style of interaction with these new technologies. Younger kids today are going to grow up where, yeah, of course, computers can understand what I say when I talk to them with nuance and sarcasm. And, of course, I have a personal AI agent that can go, you know, do my research for me for my next trip.
And of course, you know, and I think we are not even contemplating the second and third order effects like that. That led to my child thinking touching a phone to the side of my face was like, weird, um, which just still boggles my mind. But this is, I think we're at the start of a really significant trend. And I'm hopeful in a lot of ways because, you know, I, I, like many others, read things like the anxious generation, and I catch myself being, you know, mildly addicted to staring at the glowing screen in my pocket.
And you wonder if you fast forward four or five years, will software kind of melt away into the background? Uh, will a lot of things that are tedious, like waiting on hold, then you can't find a phone number, will this technology make all that go away? Yeah, yeah, call us anytime. Oh, and it knows everything about me. And whether I want to talk to it over chat because I'm on the BART train and I don't want people to hear me, or, you know, I want to talk on the phone because I'm, uh, holding things in my hands, like all of that will just be available.
Um, so I'm excited for it because I think, like all technology trends, we're in that, uh, the bottom rung of Maslow's hierarchy of needs, and it's very hard to see self-realization or whatever the top thing is. But I think we're gonna get there relatively quickly. And and I and our hope at Sierra is we can help every company in the world navigate that. You know, step one is make an amazing customer service experience for your customers that make them feel respected and valued and truly personalized.
Step two is set your company up for whatever the future holds. What does conversational commerce mean? Uh, what does it mean when people are doing their consumer research on OpenAI rather than search engines? What does it mean when, you know, uh, I'm sure many of you have done that when you get, like, a lab result, I just upload into ChatGPT immediately before I talk to my doctor, which I I don't know how he feels about that, but I promise you, 100% of his other K patients are doing too.
Like, the whole world's changing, and so, like, a lot of what we think about as Sierra is like, how do we set up every company in the world to be successful in that new world?
Technically, though, are you developing your own models, like, what is the actual tech secret sauce here that you have? Uh, is it is it models? or what is it?
We do a lot of fine-tuning. We don't pre-train any models. Uh, I think most AI applied AI companies, uh, shouldn't. It's a very fast depreciating asset and probably would not produce a meaningful return, uh, for for your for your shareholders. But it is quite complex. For any given message for to one of the agents on our platform, probably twenty plus inference calls just to generate one response. Um, just to give you a sense of the complexity.
Lots of different models under the hood. There's not one provider or even one parameter count, which is a measure of sort of the complexity of these models. I think that's the way for an applied AI company like Sierra, where the world is going, because it's almost like saying, what's the right way to store my data? And for the technologists in the room, there's like 1 trillion different databases and data storage systems, from Snowflake and Databricks to trad- traditional transactional databases.
And we've gotten to the point now where, like a modern technologist would know, hey, for this use case, this is the correct choice. That's where we're going with a, I think, in the applied AI space, not artificial general intelligence, but the applied AI space is, uh, these models are truly pieces of infrastructure. And, you know, sometimes you want something that is really fast, and sometimes you want something that's really cheap, and sometimes you want something that's really high quality.
And this price performance latency choice, there's a really an option everywhere in that matrix, uh, for whatever you want for your business. And, um, I think it will end up like the database market. It will be practitioners of, uh, building these agents and other things. They're they're not going to be the, uh, researchers who know how to pre-train a model or, uh, and I my intuition, for what it's worth, is even fine-tuning will wane over time, just as the, uh, context windows and and, uh, quality of of, uh, rules adherence improves in these models.
But there will be, uh, what it means to build an application on these models will be like, hey, do you know how to use a database? not do you know how to write a database? And those are two very different skill sets today, and I think that's kind of where the, um, applied AI market is going.
I think we saw with the release of GPT five that the models they're they're still getting better, but the step changes are not as dramatic as they used to be. Maybe that will change. The space moves faster.
I don't totally agree with you on this one, but finish your question, cause I'm rudely interrupting.
Okay, well, you're you're on the board of OpenAI, and nor nor should you. But, like, I, I guess what I'm saying is, Do you agree with the thesis that the models themselves are kind of commodifying? They're they're, they're becoming like, I mean, you you talked about it as infrastructure, but I guess what I'm getting at is, what are the second order effects if that is true? If the models are really just becoming like anything plug and play?
Yes, they have certain attributes that are better, but they're not getting dramatic, they're not dramatically step function changing like they used to.
Well, the reason I was disagreeing wasn't, uh, being an OpenAI homer, which I am, by the way, so happy to play that role. Actually, more just saying, I think it really depends on the task. Uh, if anyone was using, uh, GPT four o or four one for coding, and then swapped in GPT five for coding afterwards, you saw a dramatic improvement in performance. So, through the lens of that task, it was very much a step change in performance.
And so, you know, for people, uh, who are using this for coding agents, I think through that lens of that use case, what you said was definitely not true. Absolutely a step change in performance. I planned, you know, one of our vacations on ChatGPT earlier this year, and I I I think I was using four o to do it. And my guess if I used GPT five for that same trip planning, it would have been like, yeah, okay, it's like, whatever, you know, slightly better.
I had a great vacation, so maybe I just didn't have high enough standards. Maybe it would have gotten, like, a lot better. I think that we're getting to the point for a lot of tasks we've reached sufficient intelligence for those tasks. So when new models come out, if you're measuring it relative for planning my vacation, you'd be like, Gosh, I don't see a huge change in the quality of this model. If you're trying to discover a new therapy and you're doing drug discovery, or you're trying to autonomously write a complex piece of software, um, or you're trying to do a complex asynchronous, you know, agentic task, your perspective on how big of a step change may change.
So my my intuition, but it's just one person's intuition, is that perception of how big of a step change these models are will increasingly be a function of how complex of a problem you're trying to solve with them. And if you think about what it means to build artificial general intelligence, we need some more improvements, right? It's, uh, there was a really interesting thread on on, uh, X from, uh, OpenAI researcher who, uh, gave it a math paper, and it it actually had a relatively novel approach to for a type of math I don't understand.
So that's the limit of what I can say about that. But it was really interesting, it was really creative, it really had that sort of almost like AlphaGo moment of like, wow, that's interesting. That's sort of novel, new mathematics. Certainly, if we wanna get to the point of, uh, developing new AI research, finding new drug therapies, uh, proving some of the unproven, uh, math problems in the world, we have some work to do. We haven't gotten to that point.
But my guess is for, you know, when what motivated your comment, probably for a big bunch of tasks, the models have sort of gotten to the point of of sufficiency. And I think so, going back to your question, which is so what does it mean? I think, you know, OpenAI is a mission-driven company. Our mission is to ensure that artificial general intelligence, uh, benefits humanity, um, and and we want to work towards beneficial AGI, and we're not there yet.
You know, we need to to continue to do that research and development. Uh, there are parts of it that already are super intelligent, but there are a lot that aren't, and and that's really what we're working on. Does it mean for different tasks that Sierra solves, or you do in your personal life, we need those, you know, f- uh, really powerful models, maybe not, and I think that will just result in a ecosystem of models and and what they're used for.
But what's exciting just, you know, around here in San Francisco, like, w- we're not done yet, you know, we wanna create AGI, and that's that's really exciting. And I think, despite the perception of these models slowing down, I'm not sure I don't really subscribe to it. And and you can see in some of the true research breakthroughs, the math Olympiad results, I mean, these are really meaningful new changes that weren't possible with previous models, and I think they're they're dang exciting.
We need to take another quick break. We'll be right back.
Running a business shouldn't feel like surviving a software group project. One app for accounting, another for inventory, another for sales, and somehow none of them talk to each other. That's where Odoo comes in, an all-in-one business management software that brings every part of your business together. From sales and accounting to inventory and marketing, all in one powerful platform. No messy integrations, no bouncing between tabs, and best of all, no spreadsheets. Stop managing software and start managing your business with one unified system.
Try for free today at odoo dot com slash vox. That's o d o o dot com slash vox.
Support for the show comes from CrowdStrike. AI is already fundamentally changing how companies build products and run their businesses. If you listen to the show, you probably already know this. But while AI drives innovation, every new model and agent also creates new cybersecurity vulnerabilities. That's why CrowdStrike built AI detection and response, or AIDR. AIDR helps organizations discover where AI is being used, monitors AI activity at runtime, and helps teams detect, investigate, and respond to threats involving AI systems.
As companies adopt AI, security has to be built in from the start. More than 70% of the Fortune one hundred trust CrowdStrike to protect their businesses. Now, CrowdStrike is helping secure the AI era. You can learn more about AI detection and response at CrowdStrike dot com slash decoder.
Support for the show comes from Framer. If you wanna create a website that looks great, feels handcrafted, and comes together fast and seamlessly, then look no further than Framer. It's a complete website platform, not just a builder, so teams can launch and keep improving their sites in one place. It helps creators, teams, and businesses ship production-ready sites faster than ever, while getting every detail right. You're able to prompt, inspect, edit, and publish in one place at a whole new pace.
On Framer, agents and humans work in tandem. Agents bring speed and scale, while people bring taste, judgment, and control. And that all takes place in the same workspace where the real site is designed, managed, reviewed, and published. Thousands of businesses, from early-stage startups to Fortune five hundreds, are choosing to build their websites in Framer, where changes take minutes instead of days. Learn how you can get more out of your site from a Framer specialist, or get started building for free today at framer dot com slash decoder for 30% off a Framer Pro annual plan.
That's framer dot com slash decoder. Rules and restrictions may apply.
And we're back. I'm glad you brought up AGI. I am increasingly of the opinion that no one knows what AGI means, but I think as the chairman of OpenAI, it actually really matters what you think OpenAI means. And I would love to know, uh, what AGI means, and I would love to know what you think it means, and if that has changed at all for you. Like, in especially in the last year or so, do you have a sense of, like, this is AGI?
We we we when we have ch achieved this, we have hit AGI.
First, I'll answer the last question, which has it changed? Yes, it has changed for me. I think we are already at what I would have defined AGI at three years ago. Almo- actually, by the way, I think we're at what almost anyone in the world would have defined AGI as three years ago already. There's this thing called the Turing test that got, you know, like, which is, uh, I think, actually, I don't know what the original one was in the paper, but the way it was taught to me in computer science was, you know, having a conversation with an AI and having it being basically indistinguishable from human.
And we've been past that for years, and we passed this. This was like the big thing in AI for a long. We passed that, and just like, yeah, forget the Turing test. That was a dumb idea made by the smartest computer scientists of all time. And and so we just keep on moving the goalposts because we have, uh, exceeded our own collective expectations about what this technology can do so many times that what we had intuitively thought of as AGI, like, we've lapped it, like four or five times.
So the way I think about it now, and I I may change again, is that, uh, in the domain of digital technology and ideas, you know, is it are these models exceeding human intelligence, or at human intelligence, and in almost all all domains. I say that in the digital domain of ideas, because I think it's one thing to invent new types of math, which I think a lot of people would would put in the domain of of, uh, AGI, you know, and and superintelligence.
Interacting with the physical world is a whole different thing, and I and I think that's a separate problem that's unrelated to intelligence, uh, per se. Uh, you know, just being like, like, concrete about it. You can invent a new therapy, but a clinical trial is a completely independent process. So I think, uh, the the intelligence part of it is really what I was trying to find a measure of, and the other part of the G in AGI is generalization.
And so one of the things that I don't believe, but, you know, talk to a lot of researchers, that's what's interesting about AI, is like, some of the smartest people don't agree on all these things. But, you know, if you make something that is really, really good at math, how good it will be at a lot of other things. And you'll talk to some researchers who think, well, math is sort of the basis of reasoning, and it will be great at a lot of things.
You talk to other people like, I don't know, will it generalize to different parts of biology and other things like that? So I am more on the camp of, I think, as long as the model isn't, you know, trained for something like the math Olympiad, and that was a byproduct of the model, it will generalize. Um, but I think we'll have to see, you know, and I think we'll see how it generalizes to other forms of science and and things like that.
Uh, but more and more, just looking at the rate of progress and seeing some of the results coming out of OpenAI and and the other research labs, I'm more optimistic. Um, and I'm looking forward to the first true scientific breakthrough coming from an AGI, and I think it will happen in the next couple years. I don't know, but I I I have a i it feels like it will. Um, it's a better question for the researchers at OpenAI than me.
But certainly, if you start to see some of these early, um, early results, it certainly, um, feels possible.
Why are people like your old boss, Mark Zuckerberg, now talking about super intelligence? What is what is the difference here? I mean, this is like a thing when you're out here in S in Silicon Valley and and SF. Like, now people are saying super intelligence, and it's like, is it because we, like you said, everyone is kind of like, well, we we did it, shrug, like Turing test, and now Rebrand, it's like, Yeah, rebrand. Uh, w- what is the difference?
I don't really understand, to be honest.
Superintelligence, I think, literally just means that it is more intelligent than humans. And, uh, and so, you know, I guess if there's a subtle distinction is if you made something that was generally intelligent and functioned as well as you and me, is that, like, no lackluster? No offense, Alex, by the way, I think would be great if we made it. Yeah, uh, you're you're sufficiently intelligent for me. Um, and so I think it's, uh, a higher bar, you know, which is, you know, truly exceptional.
There's a few reasons from a research and safety standpoint, it's useful to talk about superintelligence because it's a reminder that if the models exceed your own capacity to reason, how do you monitor it? How do you make it safe? And and, you know, you really have to use technology to monitor the technology if it exceeds your own capacity to do so. There's lots of precedent in non-AI things. You have lots of, uh, things in an airplane or a car monitoring for things you don't can't understand or operating too fast.
But that is a really important area of research, so I think it's useful to talk about. There's sort of the public relations part of it that I don't really have a opinion on or or care to think about, but it is useful when you think about safety. And, uh, you know, there's a real question of, like, how do you know that it's aligned if you can't understand it, you know, and how important is it that a human being understand it versus, say, a supervisor AI that we made to understand it?
And there's a lot of both technical and, like, philosophical questions and all of that that I think are really important to answer as we develop.
Yeah, I was at a recent dinner with, uh, Sam, Sam Altman. This dinner got a lot of headlines, cause Sam said that he thinks we're in an AI bubble. And his exact quote was, Someone is going to lose a phenomenal amount of money, we don't know who, and a lot of people are going to make a phenomenal amount of money. Um, I'm wondering, do
It's like the old marketing quote. Spend 50%, if only 50% of my marketing is useful, I just don't know which fifty, you
which one?
know?
Right? Um, do you agree with that? And if so, why?
Oh, absolutely, yes. I've given this analogy before, so I apologize, Alex, if you've heard it, but I think there's a lot of parallels to the internet bubble. If you look at the internet bubble, a lot of people think about the flops like pets dot com and WebVan. Through the lens of the past thirty years, though, we've now gotten most of the largest companies in the world, including Amazon and Google, two of the largest companies in the world.
But then, uh, you look at, like, how much of Microsoft's market cap is from cloud and others, and you start to look and you say, actually, if you look at the, you know, GDP of the world, how much has actually been created or influenced by the existence of the internet? And one could argue that the all the people in, like, 1999 were kinda right, you know, kinda. It was as impactful on pretty much by every measure, and even things like Webvan.
There's now, you know, as the internet became more distributed, there's really healthy businesses that, like Instacart and and, uh, DoorDash and and others that are sort of built that are now that the smartphone and the scale of the internet has sort of matured. So even some of the specific ideas were actually not that bad, but maybe a little early. But if you look at the internet, so, like, if you, you know, were an Amazon shareholder from their IPO to now, you're looking pretty good.
If you're a WebVan shareholder, you might feel a little differently. And so both exist at the same time. And so I think right now you have just, I think, the AI, modern large language models and modern AI, um, are absolutely going to have a huge impact on the economy. If you just look at software engineering and customer service by themselves, I mean, they're just we haven't seen a world with we which we've reached a sufficient number of software engineers, and we probably will with coding agents, you know, we just cause we're the we've taken something scarce and we're making it more plentiful.
And you know, what is the market for developing software? I don't know. I don't even know how to measure that because every company in the world is now a, you know, a a a software company to some degree. So, as a consequence, it's, I think, uh, just for me, it's it's almost has to be that there's gonna be huge winners in this. And because of the amount of economic opportunity, you just end up with a ton of investors, and some companies will fail and some will succeed.
If you look at, like, the people who built out fiber in the, um, early days of the internet, a lot of them went bankrupt, but that fiber ended up getting used, you know, it just by the next person, the private equity firm or whatever that bought it. So I think it is both true that AI will transform the economy, and I think it will, uh, like the internet, create the economic, like, huge amounts of economic value in the future.
And I think we're also in a bubble, and a lot of people lose a lot of money. I think both are absolutely true at the same time, and there's a lot of historical precedent, uh, for both of those being true at the same time.
Does it worry you at all that the bubble could be in the sector of AI you're in in the enterprise? I mean, there was that MIT report that everyone's been talking about, where a lot of spend is not seeing results. I know you have a different, you know, pricing model that's more geared towards success, but, like, I don't know. The it seems like the bubble could be all the enterprises that have rushed in and are spending a ton of money on stuff that's not working.
What happens when that reverses?
I'll I'll decouple whether I worry about it and that study, which I disagree with, because I do worry about it, but I don't worry about that study. So I'll I'll decouple the two. So I'll I'll end with the study, cause it's like more optimistic than me worrying about my existential existential issues around my business. But I'll start with that. Yeah, I mean, it's weird. I so there's this like story that goes on around me about rewriting Google Maps, and and it's like, mostly true and a little embellished, like many great stories are.
And it's interesting to me because that particular thing is people like to tell the story because they're like, oh, wow, one person wrote a lot of software over a weekend, and now if you've used, like, Codex or Cloud Code, you're like, Yeah, I can just have an AI agent agent do that in a weekend. So it's like the thing that's sort of, like, so exciting that was like, actually part of my own personal identity is now literally an AI agent.
Maybe not quite yet. I I wrote some pretty good code, but, like, you know, probably in a couple years, like, yeah, an AI agent could totally do that. And so it's gonna go from wow, that was impressive, to wow, people did that, you know, like, over the next couple years. So there's the business thing, which is what is the software market of the future? I think it's a really good question because if you pull the thread and we reach plateaus, like, you know, self-driving cars, we're really excited.
It took a long time, so, you know, even smart people can be wrong on these things, or or or are too overoptimistic. But with agents doing software engineering, we're taking the scarcest resource and one of the highest paying jobs, scariest resource, and we're literally making AI agents that do that. So what will that do? I have a lot of people say, should I study computer science in school? And I have a bunch of opinions. I think the answer is yes, but honestly, no one really knows, right?
And are we gonna reach a world where generating software, um, and generating is not the hardest part of software, as most software people know, but generating software will largely become a commodity? Maybe a lot of people think that. What does that do to the software market? My hypothesis is actually, um, it doesn't change a lot. I don't think when you buy, uh, going back to my ERP, I don't know why ERP systems are on my mind th- this morning.
But, you know, if you buy an ERP system, you're not buying, like, the bits and bytes that did it. You're buying the fact that a lot of companies have their ledger on it and that you can close your books every quarter on it and it's reliable and it. They have, like, up patched their servers so that, you know, you know that your cloud-based ERP is not gonna, you know, have a security vulnerability. And they have these compliance certifications and all these other things that aren't particularly exciting, but they're kind of the boring but important part of enterprise software.
If you could write your own ERP system as a major CPG company, is that a good idea? I'm not totally convinced it is. Uh, I always like to say software's like a lawn, you have to tend to it. And so if you build it, you bought it, right? You have to own it and maintain it and deal with all the, uh, you know, there's a new accounting standard that comes out, and all of a sudden you're you have to, like, do that yourself.
So, d I think it will change the way we write software. Do I think it will completely upend the landscape of the existence of an enterprise software market? I don't totally believe that. Might be wrong. Uh, it's really new. We're just in a really new world because we're taking something scarce and making it plentiful. I've I have thought about this movie a lot recently, which is, uh, and I I recommend people watch it, which is, uh, Hidden Figures.
It's a great Disney movie about the, uh, putting people on the moon, but in particular, the women who did the math calculations to do it. And their job title was calculator. I'm a calculator. I didn't know until that movie, and I watched it with my kids, that that was a job title. And there's an interesting one of the women in there, uh, they're putting in, uh, an IBM computer, which is like the size of a living room.
And she s, in a sort of a savvy way, learns how to use punch cards to so that she doesn't, you know, ha- like, basically for job security. We're all kind of going through these moments right now, like I'm a calculator, basically. And that story of me with Google Maps is like a story of, uh, a calculator, right? And so, but I I think the second and third order effects are a little fuzzy. I believe the the enterprise software market will change towards from software to agents, but I believe companies want to buy solutions to their problems, not build software.
So I believe the market will, um, continue to exist. That study, I don't know the basis for the, um, data in that study. I think it was problematic because it conflated people building their own software with AI and buying solutions off the shelf. And I think those are two very different types of AI adoption. We have, uh, basically 100% success rate with our customers doing a proof of concept and going live with our platform.
Uh, I think it's because we're not selling AI, we're selling customer experience, and you just, like, turn it on and it works. Uh, there's an amazing company called Harvey somewhere here in San Francisco. I actually don't know where their HQ is. Really great company. I've talked to so many law firms who've deployed Harvey for a lot of their legal analysis, and they're all successful. They're, but it's not an AI project. Like Harvey uses AI, that's the only reason the company can exist.
But they're doing it because you want a better antitrust review process, and that's what they're buying from it. I think we're just in the early days of AI, which there isn't a wonderful vendor for every s problem that you have in your business yet. So, as a consequence, you either have to wait or you have to build it yourself. And so I don't know what the right answer is. I'm not, you know, and when you're trying to build these things yourself, it's just a just a glacial amount of complexity.
Uh, under and and what you end up having is a lot of these, I call it AI tourism, you know, like a lot of people doing, you know, performative AI projects. But to actually go to that last mile is is quite difficult. I think the solution will be applied AI companies. Uh, I think, you know, if you want to build a better customer experience, buy Sierra. If you want a legal AI agent, buy Harvey. And I think we need to go through every use case, from supply chain optimization to accounting to, uh, maybe an auditor for your accounting department.
All of those will be AI agents, but I think there's a company in every single one of those domains. And I I think that is the correct way to package and purchase AI software. And I think that study is a measure of all the people who are trying to, uh, which is a just lack of maturity in the market. There's just not a solution in every space yet. Um, but I there's a few VCs in the room, hopefully, thanks to all of you, in a couple years, there will be.
Uh, and I think that will be the new what was software as a service, those new agent companies will be that next frontier of of business solutions for enterprises.
All right, Brett, we have to leave it there. Thank you so much.
Thanks for having me.
All right. I'd like to thank Bret Taylor for taking the time to speak with me, and thank you for tuning in. I've really enjoyed guest hosting Decoder these last several weeks. I hope you've enjoyed the episodes. If you'd like to let us know what you thought about this show or what else you'd like us to cover, drop us a line. You can email us at decoder at the verge dot com. The team really does read every email.
We also have a TikTok and an Instagram, check those out at Decoder Pod. If you like Decoder, please share it with your friends and subscribe wherever you get your podcasts. Decoder is a production of The Verge and is part of the Vox Media Podcast Network. Our producers are Kate Cox and Nick Statt. Our editor is Ursa Wright. The Decoder music is by Breakmaster Cylinder.