OpenAI DevDay | Sam Altman, Alexander Embiricos, Andrew Ambrosino, Peter Steinberger, Tibo Sottiaux, Kath Korevec, Casey Neistat

You're watching TBPN.

Today is Tuesday, September 29th, 2026. We are live from OpenAI Dev Day 2026.

That's right.

We're very excited.

Beautiful day.

Beautiful day in San Francisco. We don't get up here nearly enough, but what a fantastic day to be here. A bunch of interesting launches, bunch of guests from OpenAI joining us. We have Sam Altman joining us in just about 10 minutes.

Bunch of friends here. Amazing opportunity. If you're here, walk in the background.

Okay.

And get a screenshot of it. We'll get a little picture together.

I see Jeff from Stripe.

over here hanging out. We're live, Jeff. We're talking about you. Great to see you. Come, come, come in the background. This is, uh, Mr. Agentic Payments over here.

That's true.

Captain Agentic Payments. Uh, we're live, Jeff. You're live. We're live. We just started. Anyways, we'll get back to the show. Yeah, there we go. Anyways, having a lot of fun here. Yeah, great group. Uh, before we get into a bunch of great guests, we should talk about the news. Anything going on?

Well, there's Ramp. Time is money. Save both. Easy to use corporate cards, bill pay, accounting, and a whole lot more all in one place. Always a good chance to tell people about RAMP. The headline on Dev Day, just to kind of set the table, 2,500 attendees. It's been 4 years in a row now. 20 launches today. I think the number might be higher depending on how you slice them up and the big numbers that they put on stage. ChatGPT is now used by more than 1.2 billion people every week. Over 35 million people build with ChatGPT, work in Codex every week. And the big launch was personal AI agents, Dots, but also ChatGPT Spaces, GPT-6.1 Sol, And a bunch of other stuff around sites, Codex, and work that we will get into. But, uh, let's take people through some of the other stories that we should cover. I'm sure we'll be covering this more tomorrow as well.

Uh, start with the journal Aura loss.

John, you want to go straight there?

Let's get straight into it.

There's so many other stories.

Some unfortunate news. The, the, the Aura IPO seems to be, uh, taking a pause or, or a, a bit of a slowdown.

Yeah.

The headline is smart ring maker Oura delays IPO due to market uncertainty. Joe Weisenthal isn't really buying it. He says the S&P 500 is up 12% this year and companies are delaying their IPOs due to market conditions. I saw some commentary on this saying that their IPO was oversubscribed by 4.4x. That sounds great. The historical average is, I think, like 2.5x, something like that. So why'd they pull it? That's confusing. I talked to some other folks who said no, more in more recent eras, Uh, you've seen IPOs go out at 10x oversubscribed, 15x oversubscribed.

I think Figma was something like 30 or 40.

It was a crazy, crazy multiple. Yeah. And also, I believe this IPO was going to be pure secondary.

And so, yeah, it's not like they need the cash.

And yeah, and so it was more of like a marketing and excitement, a key moment for the company. It wasn't a necessary financing round right now. And so, uh, they can pull back and keep building.

So on, on one hand, Amazing product.

Yeah.

Uh, have, have managed to have relatively little competition in the ring category despite much broader competition in wearables.

Yeah.

Um, it seems like it's a great business. I think investors generally are gonna be a little apprehensive to own a name that is in, in any other decade was at scale but now feels like subscale relative to some of the other action.

Yeah.

In the market. Yeah. Um, and so I'm sure they'll be back, but, uh, yeah, unfortunate news.

Are there any rumors of, of direct competition from the bigger companies like Apple or Meta?

I don't know.

This always felt like something that Apple should have bought.

Totally. Or, or competed with. I mean, a lot of times they didn't buy Pebble. They just launched the Apple Watch, but they did very well in that category.

Yeah.

But maybe it's just—

No, but in many ways, like, it feels like it would be so much— make a lot more sense than, let's say, like a Beats by Dre to just like fit in Sure. To the whole Apple ecosystem.

Oh, yeah, yeah, yeah, yeah.

Right away.

Yeah.

They could, you know, it almost doesn't need to be even reskinned, right? It feels like it fits into the ecosystem already.

Yeah.

So also, they are facing a bunch of competition from, uh, ring makers and device makers that are more AI forward. Like they have a microphone on them, but that feels like an easy—

Very unclear if people actually really need or want that yet.

Oh, true, true. You think that's why they're moving slower on it is that they—

well, I don't think— I think a lot of people, I think if they came out and said, we're adding a microphone to your ring, a lot of people would be like, you're getting— you're collecting enough of my data.

Sure.

I don't need a microphone.

Yeah, you already have my heart rate and blood.

But yeah, some other commentary here over on X says the Nasdaq hit an all-time high 7 days ago. So basically saying like, look, it's probably less market uncertainty and more of a disconnect between what investors would pay and where they want to price.

They were looking to raise as much as $2.2 billion in the IPO, which had drawn about 4 times as many orders as there were shares available. Auris is the highest-profile firm to date to delay its offering. And they list out a couple others. The stock market, meanwhile, Has broadly held up over the past few weeks despite the 10-year Treasury yields reaching their highest levels since 2007. Now it's up more today. There's a quote here from Joe Saluzzi: "I wouldn't say we're in a market that's scary, so this is a bit surprising. This isn't a market I would be scared of if I was IPOing." I can handle the market uncertainties. I don't know what they got, what they're fearing.

No, but again, like there's always there's always a price.

Yeah, clear. Yeah.

And clearly they'd rather— yeah, you know, the bankers and the team would rather sit it out and just get it done at their price when they're ready.

So Forerunner and Lifeline were going to offer 36.5 million shares. Uh, Oro is planning to sell 13.5 million shares. And so, uh, those venture funds will have to wait to recycle the capital, I suppose.

Uh, in other news, Long Lake completed their acquisition of Amex Global Business Travel.

How much did they pay?

$6.3 billion.

And they have like thousands of employees now or something?

Up to 30,000 employees across the company. Never, never have, have done a, like a RIF at any of the companies. Oh, it's actually growing headcount. Um, so again, more data that shows if you're actually applying AI well, you can actually grow, sure, uh, grow your payroll.

Sure. Uh, yeah, the other, the other big story, uh, at least in the AI world, is that AMD sets 8 billion deal for AI research company. Of course, AMD bought WorldLabs, the, uh, the NeoLab, I suppose, is the term, from Fei-Fei Li. AMD agreed to acquire artificial intelligence research firm WorldLabs, an all-stock deal valued at about $8.2 billion, an acquisition that reflects chipmaker's ambitions to control more of the AI stack. AMD, a semiconductor company that recently touched a trillion-dollar valuation, said the deal would bring leading AI research and development talent and help the company develop better AI hardware, software, and systems, especially for the growing field of physical AI and robotics. So I think—

Big moment. Yeah, big moment for Fei-Fei and the whole team. Yeah, like, I think, uh, I think it's very— I'm super excited to see what they do at AMD. I wonder how much they'll continue to focus on exactly what they've been doing to date.

I mean, once—

massive moment, massive moment for Andreessen Horowitz. Another, another uh, you know, multi-billion dollar acquisition.

I think Ev Rangel had to fill out the apology form.

Yeah, yeah, yeah, he filled out, he filled out the apology form. Credit to him. Uh, but yeah, big, big moment. Um, and yeah, it was always, uh, Fei-Fei and the team are incredibly impressive. It always felt, um, it was still, no matter how much I got explained this sort of like near-term strategy, I never understood exactly how it would translate into billions of dollars of revenue today.

I mean, yeah, people would say robotic training data, you simulate a world and then the robot can interact with it, and that's synthetic data for training humanoids or any robotic system. The more near-term one was you could sort of see the progression of world models to a point where you could create a video game on the fly and have all of the generation and all of world creation happen on the fly with just a new technology. And you would see, you know, companies potentially license that technology like people pay Unreal Engine to, you know, host or develop their games on top of. There was something that was happening there with consistency. Google has a product as well. But yeah, they were— there still hasn't been that like ChatGPT moment for adoption of world models, really.

Yeah. And you can imagine this is like a, like a future-facing bet.

Yeah.

As in, hey, let's make a much bigger play into, into robotics.

Yeah. Yeah.

Yeah.

I mean, Jensen's been doing some more stuff with world models.

Here he is.

Open source. We have Sam Altman joining for the 3rd time. Welcome to the show.

Man of the hour. Good to see you.

Congratulations. We're going to have you throw that headset on. Welcome to the show. Thanks for having me. How is Dev Day 2026 going?

People seem very happy.

I think they are. What were you doing 20 years ago?

Uh, God knows if you—

no, I was like working on a little startup.

Didn't Loop launch 20 years ago, basically today?

That's incredible if that's accurate.

Yeah, I think September of 2006 was the launch. And I was thinking about it because, uh, I, you know, as we enter the moment of personal agents, like a lot of— there's obviously work applications, but a lot of the demos that I've seen people talk about are things that feel aligned to the original Looped vision. It's like, if my friends are in town, let's plan a barbecue, let's order the food, let's go to the restaurant, let's get the reservation. I'm wondering, is this a full circle moment for you?

It's interesting. It didn't feel like that until right when you said it.

Okay.

Now I guess it sort of is.

Yeah.

And I'm wondering how you have been using Dots, how you think this will actually instantiate itself into people's lives. It's so easy to lean into the work side, but this feels like a reinvigorating moment for OpenAI as a consumer company.

Look, sadly, all I do these days is work. So I've mostly been using it for— and I hang out with my family. That's really the only two. But I have mostly been using it for work. And it is amazing for that. And I feel like it has gotten me back, like, my attention and time and ability to, like, explore and do new things. And I love it for that. I have not yet figured out how to, like, go use this to have more fun in my personal life, but I'm inspired to go do that.

But you're optimistic that people will ultimately do that. I feel like, you know, OpenAI has been called, like, the accidental consumer company. And at the same time, time, the vision seems to be like actually get AI in the hands of people and maybe let's not end up in a future where that means everyone working 24/7, right?

100%. I, you know, I think the world's always confused when we talk about AI about whether we want— we're worried people are going to have to work too much or too little.

Yeah.

But I think it's great to give people like more time to hang out with their kids, do their hobbies, work on creative new projects. I think most people actually want that. And I really feel it with DOTS.

Yeah.

So I, yeah, I think people are gonna get that.

Yeah. What has the last— I mean, the last time we talked was a couple months ago, or maybe more than 6 months ago on the show. Uh, it's been a wild time. Like, a lot of things have gone right, it seems.

Yeah, holy shit.

Crazy.

There's some numbers that leaked today, but it seems like some things changed. How do you tell the story of the last 6 months or so?

Um, I mean, I think the answer is it's all just one jagged exponential that's going and probably the next 6 months will be even crazier.

Yeah.

Um, the, the technology has been capable for a while. It's clearly gotten much more capable, but it takes a while for technology like this to diffuse through society. And I think in the last 6 months, people are using it much more. And now with these sort of always-on agents, I think people will again go through a period where they're like, wow, I can really use this in all kinds of ways. And this is like a significant new thing in my life that's helping me.

Yeah. So is this a moment to sort of like pay down the capability overhang right now?

You know, when you were talking about how, you know, you can use this to like do the stuff that we were thinking about 20 years ago and like book a restaurant and, you know, coordinate with my friends. I was kind of like laughing in my head because I was like, yes, you can do that, but it can also like solve the greatest unsolved math problems. And so the delta in capability between, you know, some of the stuff we're talking about now and what these models are so smart can do is just going to take us a while to figure all that out.

Yeah. What have you learned this year about the sequence of events? I was shocked by if you had told me that there was going to be Hugging Face and then Navier-Stokes and then After that, I saw the first person said, I used ChatGPT to book me a haircut. I would have not predicted that those things happened in that order. Like, I maybe would have predicted that they all happen in the future.

Yeah.

But I would have predicted we get the haircut first. And, and I'm wondering if you've learned anything about not just predicting the future, but the sequence of events.

I mean, one of the obvious things is just people want their lives to be a little easier.

Yeah.

And actually, I think more people are interested in just their lives being a little bit less hectic, a little easier than they are in whatever Navier-Stokes can do for them, or at least their direct experience.

Totally.

So, um, but I think the cool thing about this is that models that are this capable, uh, can— are going to be applied to make people's lives better in big and small ways, in ways that we can't even really imagine today. And, you know, I think the fact that we all talk about haircuts and booking haircuts is like a It's a failure of imagination, but the way it gets corrected is all these people here.

Sure.

Like, as I've had a little bit of time today to talk to what people are building here, like, I am amazed at the creative spirit of what people are building for each other with this technology. And I think it's been— I think it's been one of the greatest surprises of the last 6 months is seeing, like, just how far people are pushing the developer tools to build this amazing new stuff.

Yeah.

Talk about some of the new functionality, being able to log in with ChatGPT and basically bring compute with you anywhere across the internet. Because to me, that was— there was a bunch of great moments earlier. But to me, the moment that stood out is imagining as an independent developer that has been sort of held back by their ability to scale cool products because, hey, I can't just start losing $20,000 a day serving free users.

I think it's so important to build an open ecosystem here. And we want people to take— well, to bring whatever they want into ChatGPT, run other apps there, but also to take their subscription and go use it elsewhere. And any harness you want, any service you want. Like, this is— I have been disappointed to see other companies with AI subscriptions go in the other direction. Like, this is a thing you buy and you should be able to take it with you anywhere. And you should be able to go use all of the wonderful things these people are building. And I think we're going to push that much further.

How do you think about the potential transition from OpenAI's products being very pull-focused? As I mean, like, The app is there unless I come to it with a question, with a prompt. It doesn't do anything to a world where my dot might message me. And then you get into how much is it messaging me? How much is it pulling me in? What are the retention? There's, there's a lot of potential risks with that, right?

It, it, yeah. I mean, there's clearly ways that people could misuse or growth hack the fact that AIs can I'm sure we'll see some companies get that wrong. We're going to try very hard to not do that. But I— my own experience over the last few weeks is I actually want more. Like, I think we've tuned the dial a little too conservatively for now. And the fact that everything my Dot sends me is something I want means I think there's something still left on the table. But I can easily see how this goes wrong in the other direction, and we'll have to figure out how we're going to navigate that.

Yeah. Open ChatGPT and realize there's like a very important email that I need to respond to. And like, I didn't— it somehow like got mixed up in— and then you're like, you kind of wish it pinged you, right?

Yeah.

What can you say? Any advice for, let's say, public company CEOs that want to maintain a startup culture at scale? Because that's something like, you know, OpenAI is still a private company, but the impression of seeing how this company works is that It's not just like, yeah, we act like a startup at scale. Like, it's actually like that. It's like incredibly scrappy. It's very small groups of people trying to do the impossible on impossible timelines. And something that you're doing is working because it allows this like incredible velocity. It's not just the models.

Someone said to me a few days ago that joined OpenAI somewhat recently and had been at a big kind of slowish public company before. Was that the key is to have the work and expectations per person go up over time. And the way that big companies go wrong is there's too many people with not enough to do, and they kind of like fight each other over like these fake levels of work. And if everybody is really busy with real impact and there's the amount of work per person and stuff to do per person keeps increasing, then you can fight the like slow drag and politics of a big company.

On the flip side of that, I feel like there's a lot of companies that are going through sort of a sloppification and they're fighting slop grenades. I've wondered if the, like, the correct KPI is maybe we should all be producing the same amount of words in our emails.

Yeah.

And it's okay if AI is involved, but you don't just want to take every text message and turn it into a 20-page document, right?

We fight super hard against the slop grenade.

Yeah.

And the many slop grenades. And I think just like very strong cultural norms that if you are writing something and then you are having ChatGPT expand it and then somebody else is going to have ChatGPT summarize it back to your bullet points, the right answer is to just send the bullet points.

Yeah. I've told some people, just send me the prompt and I can expand it myself in my mind sometimes. How are you thinking about compute these days? There's so many different you know, battles and different markets across consumer and enterprise? What is the plan for the next couple of years?

To build a lot more compute.

Okay.

One thing that OpenAI has that doesn't get as much attention is our chip program, which will start to come online in the first half of 2027 and will scale a lot in future years. I think gives us a huge inference advantage for whatever, for whatever like watts we can find in the world. And given the growth and demand we are seeing across consumer, developer, enterprise, really everywhere, And to say nothing of like, people are going to love DOTS and then they're going to say, well, can I have 10 DOTS or 20 DOTS or 100 DOTS or whatever? We will have to make more efficient models, more efficient chips, and figure out how to build a lot more infrastructure.

How do you see the user experience across the different models, the different cost and speed parameters being vended to the actual end user? Is that something you want to maintain forever because developers love hearing about 6.1 Sol in ultra fast mode? Or are there consumers, you know, I noticed DOTS doesn't have a model picker.

I think people are really tired of the model picker and the like model, you know, what it like this model, that model. Like, like we want to find a way to really simplify that. Um, like the dream is that there is a single model and it kind of makes great decisions for you. Uh, and you can maybe turn a speed dial when you really need it and a cost dial when you don't. But, you know, we were talking about like models that can book haircuts versus solve Navier-Stokes. I think this model should be able to figure out a lot of the tradeoffs and we should be able to do a good job with routing. I think we'll always offer choice to developers.

Yeah, sometimes you want to drive a Ferrari to the grocery store.

Sometimes you do. So I think we'll always offer that choice, but most of the time you don't. You know, most of the time it's the minivan to the grocery store and the Ferrari to the track. And the model should like be able to make that smoother.

Yeah.

Do you feel like you have more or less clarity around the future than you did a decade ago? Because when, again, I always, I always encourage people to look back at, at your blog in the sort of 2015, 2016 era, because there was a bunch of things that you and, and other people that were writing at the time just like pretty much nailed entirely. And I go through life now knowing that there's people out there that have like very accurate predictions around the future, and sometimes there's an incentive to share, sometimes there isn't. But how do you think about your own forecasting ability?

It's nice of you to say that. There's plenty of things I was wrong about too. I try to leave the wrong predictions up, obviously. But I think the important point is you get some things right, some things wrong. I think we'll be very right about some things. I think we'll be very right that intelligence continues. I think we'll be very right about the rate of scientific scientific discovery. I think we'll be very right about how people kind of are going to have the same things that they care about and drives in their lives 10 years from now they have from today. And even with this amazing intelligence, a lot of parts of life are going to stay the same. But then a lot of the details and kind of what order science goes in or whether it's haircuts or Hugging Face or Navier-Stokes first, I'm sure we'll be wrong about.

It feels like that. Leadership skill, or basically having the mentality of a venture investor and somebody that's made a bunch of very long-term bets makes people very well suited for today in the sense that having that openness to, if I get 3 things out of 10 right and I get them really right, then that's great. Like, it sort of pays for all the other bets.

One of the things that I think is coolest about this new generation of people building with AI is you can try something so quickly and so cheaply that if you have— if you are a developer and you have 30 ideas, the canonical advice that was correct 2 years ago is pick one and put all of your effort into it because you can't build 30 things. You can now build all 30 things and you can get people to use them all and then you can work more on the ones that work. So I think this mindset of lots of ideas, quick feedback, and be open to a lot of things not working is great for this new era of how people build.

I have one more question, then we want to get you to hit the gong. On alignment, you've been deep in the weeds, I'm sure. There was a debate I saw. Is it an engineering problem? Is it a science problem? Is there a different frame of mind that we should be approaching alignment with?

I think it's like all of those things, but it's certainly incorrect to talk about this only as an engineering problem. Anyone who says we have solved the science of alignment, I believe, is wrong in a very dangerous way. Like, we need to make more research progress. I assume we will. We've been great at making this research progress. And of course, there's a lot of engineering work to do too. You know, we need to continue to figure out how to build better sandboxes and better monitoring tools. But eventually, as we— if we are going to create models that are like much, much smarter than all of us, we have to actually solve the science of alignment.

Have we considered moving away from the term sandbox? Because I have a 5-year-old. If I put him in a sandbox, he's not going to stay there. I think we might need a new term.

My kid, unfortunately, will just stay in the sandbox. Oh, okay.

Okay. Yeah, but, but, you know, different alignment.

But yes, no, I think the models are—

I think hold this up, John.

Okay.

How do I do this?

Just however you want.

Just with—

I just like bigger. Yeah, there you go. There you go. With authority.

Thanks for having me on.

We'd love for you to sign this as big as possible. Big signature right there on the gong. Thank you. This is going in the Museum of Business at the TBPN UltraDome. Thank you so much for coming on the show.

Great to see you.

Congratulations on a great Dev Day. Uh, we will move on to our next guest. We have a few more hopping on the show. We have Alexander Amberikos jumping on in just a minute. Uh, we will see—

A legend.

With him on all things Codex. We need a slop grenade button.

Being able to—

Who said that? All right. That's banned in the chat. That's great.

We could sort of simulate that if we have a, a remote, you know, remote guest on. We could say, hey, could you check, check out, uh, I have an idea. I have an idea. I have an idea for you. What do you think of this? And then you throw like a, 6-page PDF.

Oh, okay. I was thinking a slop grenade would be like you go full back to Nickelodeon slime, and when you push the button, it just slams Tyler with it, and he just keeps covered in slop. I think that's the one we gotta go.

I don't know if you—

Oh, wow.

Look at, uh, look over here. You got a crowd chasing Sam around.

They want more tokens.

Gotta be careful with this signed gong.

Yes. Oh yeah, that's a hot commodity.

Uh, over in Washington, CNBC reporter was trying and failing to interview tech CEOs at the White House. Alex Karp just took the microphone out of his hand and started asking them his own questions.

Interesting.

No, he's literally— they're doing like a man-on-the-street interview and just Karp took the wheel.

Yeah, that's great. Well, there's more stories. We'll get to those in just a few minutes. But let's bring on Alexander. Welcome to the show. We're going to have you throw that. Headset on. Do you need anything to drink? This is, uh, water. Okay, guys, well, welcome back to the show. Always great to see you. Uh, how is this Dev Day for you?

Uh, I mean, apart from the outage, excellent.

Okay. Yeah, yeah.

Wait, outage? Uh, global outage going on, or—

Or yeah, yeah, we had a bit of an outage, uh, during the demos. Like, you know, I haven't like deeply read up on what exactly happened, but like because we launched 6 6.1 Sol and just immediately got hammered.

Yeah.

Uh, we did that.

Oh yeah. That's a, that's a lesson. That's like, uh, the Meta, Meta Connect last year. Boz accidentally set off like the, the key, the, the keyword.

He just said like, what's the wake word?

The wake word. But every device in the whole space started going off and then crashed everything. So yeah.

Positive framing is a champagne problem, but broadly this Dev Day, I'm so excited.

Like, okay.

When I joined the company like a little over 2 years ago, like this moment is the moment I've been looking forward to the entire time.

Okay.

You know, like we had this chat. We invented that paradigm. It was awesome, informational, couldn't do things for you. Okay, then we had coding agents, agents that could do things. Awesome. That's the next level. But you still have to like really babysit it and figure out like, what am I going to ask this agent to do?

Yeah.

And so we're finally at the stage where I can go to someone and say, hey, if you want to do something manually, do it with Codex. Yep. If you want to delegate it, sure, go use a dot.

Yeah.

Right. And this dot, it has your back. It figures out how to help you. It just helps you get so much more value out of this. So for me, like I've been, I've been looking forward to this for years.

Yeah. Does it feel like—

Yeah, it's the same.

It's the same.

Sorry to jump in, but it's the same. dynamic of like, if you had a bunch of, if you basically had a bunch of people on your team that would just sort of flag information, but then not take any action. They're like, hey, this thing, it's happening. You're like, okay, like, why don't you do something about it?

Yeah, listen, I like that, but I would actually say like Codex is actually more like a principal engineer, but who doesn't like initially, unless you tell them exactly to do it, doesn't read like Linear or like doesn't check Slack. And so the super smart principal engineer, but you know, you have to make sure you sort of give it the right instructions.

Yeah, yeah, yeah.

Whereas a Dot, it's like you have that principal engineer because your Dot has its own computer. It can do anything that you can do on a computer in the cloud. It can also connect to your computer. So if you have a really awesome local codec setup, it can actually go use that. Except that it has its own identity in Slack. And so it's available wherever you talk. And, you know, I don't know if you guys have seen this, but it's— or for anyone listening, the way that Dots show up in Slack isn't just like a random name like Wisp. We tried that. We had agents with random names and no one knew what was going on. It was a bit confusing. But they show up with like your username. So like I'm AE, it's like AE-dot in Slack. And so you have all these, these people and their delegates just doing stuff together.

Yeah.

Uh, you know, principal engineer level capability.

Yeah.

Or, or if you're not an engineer, just like, you know, maximum capability.

Yeah. I mean, the, the demo that I saw, uh, just randomly through a friend was like, we were looking for Sam Altman's headshot and we were able to delegate that to a dot that was able to go find like a file in a file system.

That's easy.

But just that little retrieval would be a couple clicks on a web browser and now It's just—

Not even a couple clicks, it's like searching around folders.

Sometimes it's like, yeah, digging through stuff. And so, yeah, it feels like we're hitting this inflection point of like leaving just the developer workflows and getting into like everyone's workflows.

Totally.

And you always had the pro, like the prosumer of the enterprise who was like, oh yeah, they're a sales rep, but they also use Codex. And now it's like, there is a product that's approachable for everyone. Are you, it feels like there's the expansion of territory that's happening within businesses. And then there's also like different types of businesses that are able to actually get a great value. Started with software development firms and software companies. Are video game developers the next leg up? Like, who are you seeing here at Dev Day that you're like, oh, I haven't actually talked to that many of your— of that many folks in your industry, and maybe the progress in Astra is flowing through? I mean, that's what my experience was. I was modeling stuff in Blender. I haven't touched Uh, 3D rendering for a long time, and I can— I'm building video games and stuff. And, um, and I saw that, okay, this is— we've crossed this— the spike of intelligence has spiked beyond what's relevant in this particular category. And I'm wondering where else you're seeing it or what's been the most visceral to you.

So I agree that like, I think for video games, like the stuff we've been seeing people make with Astra. Yeah, absolutely awesome. And so therefore, because DOTS run on Astra, we've just seen people having Dots just like autonomously building like really fun games and stuff. So that's, that's awesome. And I agree with that. But actually I think the, the audience that I'm the most excited for to like get even more capability is maybe like solopreneurs or like small business owners. Totally. Yeah. Actually, because you know, that's where you're, you're doing absolutely everything that your business needs. And so you're just like swamped with all this stuff incoming. Yeah. Um, and that's where having a Dot that just like has your back and is like managing all this stuff for you.

Yeah.

Doing the work that maybe you could have done before with Astro, but it was just like a lot of work to figure out that that work even needed doing.

Yeah.

Is helpful. So I've heard, you know, obviously a lot of great anecdotes from, from engineers, like people waking up in the morning to kudos from their team because their team was collaborating with a dot and it helped them out. It's like, good job for having a dot that did this.

Yeah, yeah, yeah.

Which is so fun.

Yeah.

But, um, you know, we also have a lot of great anecdotes from, from like people in completely other areas. So like, you know, running the launch, like dots going around realizing that, oh, this thing about the product changed, so we need to go update the blog or like You know, your travel plans for this conference have changed, so let me just like organize things for you.

Yeah.

Uh, walk me through the specialist dots. Like, is that more of like a mental framing for the user to let them know that this capability is there? Is there actual like the harness matters narrative where you do want to go to that particular dot? Or is it because I can imagine as a solo entrepreneur, uh, it's just valuable to say, okay, I at least have like one arrow in the quiver for legal, one arrow in the quiver for finance, etc. But also, these models are pretty general, and the G in AGI sort of leads you to throw, you know, Codex, Astra, Ultra at every problem. So how do you think about coaching people through that dynamic?

Yeah, I mean, so there's 2 things going on here. So we've got DOTS, yeah, and we've got specialist DOTS. And the reason we're leading with dots first is like just very related to our mission, right? Deliver benefits of AGI to all humanity. And, you know, when I think about product, I want that to be palpable. Like, I want every person to have direct access to something that is empowering them, extending them, helping them move faster. And so those are dots. And those dots, you know, each person has one. You'll have more in the future. And it works, you know, extending you. So using your connectors, for instance, right?

Yeah.

So that's great. But when you think about how you, how you like scale the amount to which this capability is deployed within enterprise, there's actually a shift that we're starting to see happening that these are a part of, right? One of those shifts is like moving from local work on someone's computer to moving work in the cloud, right? And so specialized DOTS have their own computer, just like DOTS. The other shift is actually moving away from working with like an individual's identity to moving with an IT-managed, centrally governed, well-controlled identity. And we actually had an announcement going out today about how we're working with Microsoft on that as well.

Yep.

And so if you are a team or an organization And you have a specific type of like functional work that you would like to like accelerate the whole team with. Yeah, that's where you deploy a specialist bot. So it could be something like procurement or accounting. Those are some of the early places we've seen a lot of success. Yeah, but ultimately broadly we see that it'll be useful in many more functions. Yeah, and departments.

Personally, I feel like, I don't know, maybe I'm just like too much of a normie, but, uh, I feel like this is the frame of mind that I want to be in. Obviously there's like the merge where you the AI becomes you. And if I want to send you a message, it comes from me, but the AI is like puppeteering me. And I actually prefer the idea of like, I have an assistant and they are their own entity and they have their own phone number and email eventually, and I can share information with them and they are on my team, but I am still me. And I think that that is something that I would bet on that being relatable to a lot of people.

Totally.

We'll see. I don't know how.

Yeah. I actually made some like pretty funny mistakes as we were building this product that we had to like pull back.

Yeah.

And so like, okay, I totally agree with what you said. And then I'll tell you about them.

Someone like asks your dot, what is, what's he writing about me from Performance Report?

Well, no, that actually, there's a lot of layers of defense that's not going to leak that information. But no, it's more like what we've realized is that, again, there's this kind of like this bifurcation, right? Like, I want my dots to augment me. And so maybe I want to collaborate on an email and then I'm good. I just did that from within the ChatGPT app and I hit send, right? And that should just look like me. But, you know, I approve that email to be sent. And I made a mistake about that. Well, I'll tell you about. And then on the other hand, you know, we had, we had for a while all these agents just sort of acting sort of as us, sending messages as us with this like sent by ChatGPT thing internally. And it was like fine, but it kind of like started to get noisy. And so that's when we kind of iterated our way towards, no, we're gonna have like a strongly separate identity. They're gonna look different. We've got these characters. We're gonna have like a name that like clearly is like owner-bot.

Yeah.

Right. And that just like clarified everything that was going on in communication.

Totally.

So it's awesome.

And I love, I love the idea of like, look, sometimes the amount of back and forth and compute that's going to be dedicated to some negotiation, if we're trying to find even just a time to get lunch, they might be— our dots might be going back for a long time, but it's not going to ever be like you opened your messages to me and it's a flood of messages because they were going back and forth. It happened below the fold. And then you can, you know, you can abstract that away as much as you need to.

Agreed.

Very cool.

I'm going to set up a dot where every time I get like a startup pitch deck that's just clearly AI generated, it'll generate a playable video game. And I can just go through, go through the deck in the form of like, you know.

Certainly possible.

Yeah, but that's totally a thing you could do.

No, I know. Like, I'm joking, but it's actually—

So here's an interesting thing about, about space. Right, and sites, which is like a lot of the tools that we use to do work today are optimized for human authoring. Okay, so what I mean by that is like, you know, like think documents, right? Like you go into a document and it's like you can type into it, right? And the way that your document can look is like pretty simple. It's like text.

Yeah.

Right? And like, yeah, various document editors have like fancy ways for you to go in and like move the images around and stuff, but it all generally kind of sucks.

Yeah.

Because the tool is well designed for humans. Like, I think those tools are good.

Yeah.

Even though they're a little inflexible. But now that we have DOTS and we're collaborating with them, we have AI, like HTML authoring is like basically free. Yeah, right. And so like I love how Space incorporates like HTML visualization.

Sure.

So you could just be like, hey, like build me this artifact that I can click around or like literally put a game in the doc or whatever. Yeah, in the page. Um, and it's super interactive. And I think this is like sort of the paradigm shift that we're starting to see.

Yeah.

How do you think about, uh, the road to and the benefits of actual direct integrations? Because At the end of the day, like, if I have some service that's, that I want to interact with, like, there are APIs, there's ways, there's computer use at this point. Like, everything is somewhat connected. What's the benefit of actually directly engaging with OpenAI, doing a proper connection? What are the benefits? What are the costs?

Well, I mean, there's a few layers here. I think it starts with, yeah, now that our agents can use the same tools that we use, i.e., computers, then there's kind of this baseline, right? Like, your agent, if you wanted to and allow it to, it can go to anything. And I do want to emphasize that like DOTS have many layers of like safety protections and one of them is custom rules, right? So if you're a company and you want to limit what exactly DOTS can do with various tools, like there's a very easy affordance for you to go in and do that.

Yeah.

But okay, so that's your baseline, right? The next layer that I think we could talk about is like connectors. And mostly it's just like it makes things faster. It lets your product be a little bit more opinionated.

Yeah.

Just improves the experience for everyone.

Yeah.

The other sort of form of integration that I'm not sure if this is what you're getting at, but that I'm really excited about is plugin extensions.

Yeah.

So, you know, we've long had this idea that like you— or there's actually a fun debate in this industry, right? Which is like, to what extent should UI be generated just in time for the user?

Yes.

And we're doing a lot of interesting work here. And there's like some cool stuff coming out soon around that. But my opinion is that for UI that I'm going to click around in, I generally want it to be somewhat predictable. So I know where the button is. I can achieve mastery over this tool. But we don't want to be the only people deciding that. We want other people to be able to build their own UIs for our users. And so something super cool is that now if you're a ChatGPT user, you can install a plugin that is a plugin extension that actually renders like custom UI, custom pages just in the ChatGPT app. So if you're a developer listening to this, you have the ability to go reach your customers directly. And the meetings plugin that we just launched today is an example of this. It's like a fully arm's length plugin integration.

Yeah. Then what I— so I feel the same way about on-the-fly UI generation, custom UI. I'm excited about that. At the same time, I can sit here and talk to somebody who's extremely bullish on voice interfaces, and they think that everything should be vended over a conversational voice interface. And I feel like OpenAI has kept enough irons in the fire across all these different paradigms. Is there a tension there? Is that deliberate? Is there an idea that to reach everyone, to hit 1.2 billion MAU or WOW, you don't want to actually go all in and narrow? You need to meet people where they are across multiple modalities?

Yeah, I would say so. Like, there's something Thibault said earlier today. It's like, at some point it's about the active intelligence and the interface stops mattering. Yeah. And, you know, my, my like slight way of reframing that would be like, it's like the interface should just be whatever you want. Yeah, right. So like, I'm a product manager, so I spend a lot of my day in Slack. And so I mostly talk to my dot in Slack.

Oh, sure.

Right.

Yeah.

But then when I'm commuting, yeah, like I was coming in today, I knew I was gonna have a bunch of conversations. I was just talking to my dot like about the day and what I just prepared for was like grilling me on some hard questions.

Yeah.

You know, um, Holly earlier at the demo, the keynote today, was talking about how like actually the anecdote was kind of me. Like I saw her in this phone booth all the time and I was going to give her some feedback. I'm like, why are you in so many meetings? Like shouldn't we be like not in meetings?

Yeah.

And then she's like, no, I'm talking to my dot. And I realized this and I'm like, wow, this is crazy. We've like unearthed a completely new paradigm.

Yeah.

So going back to your thing about irons in the fire, like I think that some of these things are kind of emergent. Like I I really love talking to my Dot over voice. But the way that that happened is the voice team that was working on Dots did an amazing job. But it was kind of like we combined these 2 capabilities. And because both irons were very hot, they combined really well. But if you had started from scratch, I don't think we would have gotten to such a great outcome.

Yeah.

Yeah.

I've noticed that I don't think people here have this problem. But, uh, as I was thinking about, you remember this viral video, this guy Jim from Charleston AI, did you see this?

Ooh, he might have to refresh my memory.

Basically, he has a retail store in Charleston.

Physical store that you can walk in?

You can walk in and he'll teach you like how to use Codex and ChatGPT work.

Yeah.

And I've just noticed this thing where normal people outside of the tech industry don't just know that you can ask the AI, can you do this thing?

Yeah.

And it's like, Like, it feels like as there's 20 new launches today and it feels like the biggest bottleneck is like effective product marketing to the sort of long tail of users on how to use these different tools. Because when somebody asks me, hey, can I do this thing with ChatGPT? I say, well, have you asked ChatGPT that? Like, it'll sort of explain the general capability set and then maybe like even prompt you on how to prompt it, et cetera.

Yeah.

But it feels like outside of San Francisco, that's like the the biggest bottleneck to like adoption and diffusion is just people not realizing that the best teacher for the thing is not even necessarily a YouTube video about it. It's like being able to go back and forth.

Well, I mean, 2 things on that. First, thanks for having me on the show.

You know, this is fun.

The second point though, I think is that at least when I hear that, you know, I think this is actually a product problem as well.

Yeah.

Like if the product, I mean, it's so powerful, right? These open-ended sort of chat interfaces, you can ask whatever you want, but it's It's hard to realize that you can try that. Like, I find that very understandable. And so that's part of why I'm so excited about DOTS in this era of active intelligence is that, well, maybe I don't have to realize that I can ask for help for finding the headshot.

Yeah.

You know, it'll just notice that someone was urgently asking me for the headshot.

Yep.

And so it might just go ahead and proactively research like, where is the headshot? And then come to me like, hey, someone asked you for the headshot. I happen to know where the headshot is already. Do you want me to send it? Right. And so I think we're about to achieve this like step function change in how much value people are getting from existing models. Like God is built on GPT-6 Astra. And yet it's going to unlock so much value for people.

Yeah, yeah, I know. It's an exciting time. Thank you so much for coming on the show.

Thanks for having me, dude.

Have a great rest of your Dev Day. And we will move on to our next guest. Andrew Ambrosino is coming on from OpenAI. We will bring him on, have him throw on this headset and introduce himself because I believe this is your first time on the show. This is my first time on the show.

Welcome.

Welcome.

Good to meet you guys. Thank you so much for hopping on. Give us a little tour of your day-to-day, your role.

Introduce yourself for everyone. So since the original Codex app, I've led that team who runs desktop from the Codex app to now. And I've also picked up leading the space team.

Okay.

From the space product.

You needed some more on your plate.

Yeah.

Wasn't enough, man.

Desktop was solved. Give us an overview of your experience with the merge, putting everything together.

There's been so many updates.

hard decisions, so many Band-Aids to rip off, I feel like. What's it actually been like? What's gone well? What do you wish you could redo?

This is my merge shirt.

Okay.

Yeah.

Maybe introduce the idea of the merge because there is that Sam Altman blog post from a decade ago about the merge, and that's a different thing.

That's a different thing. Um, well, there's been a series of merges.

Okay. Should we merge them all together? We should merge the merges.

We split, we merge.

Do that eventually.

Um, no, I mean, we had this great app with Codex and we had ChatGPT that served a billion users.

Yep.

And, um, we needed to find a way to join these 2 worlds and bring the agentic side of Codex in this app into ChatGPT with Chatwork. Of course, when we launched it originally, there were still like a lot of things that didn't match up. And so we've been working on merging those. We had 2 great merges today, um, that are a little bit under the radar actually.

Okay.

Uh, one of them is that we completely rewrote chatgpt.com in the last 5 weeks.

Okay.

To be the Codex app.

Wait, your pinned tweet.

Yes.

It says, I'm going dark for 6 weeks or something like that, right? That's what this was?

Yeah.

Okay, now we know.

So now we've got the same Codex desktop experience you're used to, the ChatGPT desktop experience. It's now chatgpt.com.

Okay, got it.

And then the second merge that we've done is now by default, ChatGPT work on the desktop works on the cloud and still has access to your local machine. So you've got these threads everywhere. Codex Cloud we launched today.

Well, what that means is like by default you will be able to do something that needs a file system, that needs some local-ish compute, even though it's in the cloud.

Yes.

You won't have that fit— you won't have that tension between like, am I just in a ChatGPT thread that's sort of like short-lived versus I gotta go to my computer. Now by default you have more cloud access.

Yes. One of the things that we really like about all of our new products today, whether that's the Dots, the Codex Cloud, ChatGPT Work, all of this stuff is running in the cloud and has access to your local machine if it needs it. So if you've got files on your local machine, if you're logged into your browsers on your local machine, the whole system. We are working on this merge. And we are next working on the toggle merges.

So question for you. How do you balance trusting your own instinct around product decisions? Because obviously you would never make a big product decision that you thought was going to lead to a worse experience with real-time user feedback around how things were because people You know, we were just talking with Andrew, like, the people build up, like, a habit around different flows and different ways that the product can be oriented, and then they sort of get— build an attachment to it. And so even if it needs to change, it can be jarring. But you guys are sort of, like, drinking from the fire hose of user feedback every single day. And whether you're taking, like, going offline for 5 or 6 weeks or whatever you did, I'm sure you're getting, like, tags and comments and messages constantly being like, do this, do that, do this, do that.

Lots of tags, yeah. I think when it comes down to it, we want to put these capabilities into people's hands. And so it becomes obvious that even if for whatever reason we can't make it, you know, we can't bring chat and work together quite yet, what we knew was that what we were getting out of Codex had to be brought to everybody. And now we feel similarly with Space, with DOTS, all of this stuff. We try to put the capabilities into people's hands. We dogfood it extensively internally. So we've got thousands of people who rely on this stuff to do every part of their job. So we get good feedback loop there. But it's all a work in progress. The stuff is moving incredibly quickly, as you know. Lots of tags on X all the time. And we love that.

Can you walk me through how there's this culture of, like, side project— side quest was the buzzword. Sometimes they get killed off.

No more side quests.

No more side quests. brought back? Like, like all the computer use stuff that's going on, huge like leap forward out of nowhere. But you go back 2 years ago, I remember Operator, right? That was like, was it a side quest or was it really important work to do because it sets you up for this moment? And I'm wondering if there's prehistory for, uh, any of the new products today, like Space or anything else that you've touched on where you've like, oh yeah, like actually we, we, we revisited some work we did previously in a product It wasn't quite there. The model got better. We were ready to launch.

I think most of the stuff that we released today is like a take 5.

Yeah.

Even just of things that we did internally.

Yeah.

There were so many variations of DOTS that we saw internally, played around with. Some were local, some were cloud. Like, there were so many variations of the stuff that we used. Space has gone through a number of iterations. We had part of chat called Canvas that we had up for a while.

That's right.

We've had a lot of internal knowledge base products. So we've gone through the revs on that. Codex Cloud is obviously a full 2.0. The original Codex was cloud, then we got into a local agent era, and now we're back putting the full local agent on cloud environments. So I think everything that you see today is one of those operator stories or that operator to Atlas to Codex, right?

Yeah.

Sometimes I think that the models are changing so fast that it's hard to get the product timing on these things. exactly right to the model when the model is best at it.

Yeah.

So sometimes we're, you know, slightly too early, slightly right. And, um, but I think like the stuff that we have out today feels, feels magical.

Yeah. Uh, so tell us more about Space. What is the vision, especially for where that product goes, how it fits into, um, like a, just a typical person's workflow in the modern era?

Yeah, we, we find ourselves ourselves bouncing around between applications all day, right? We're, you know, in this, in the Codex desktop app, ChatGPT desktop app, um, all sorts of documents. We're sharing things across Slack. And so we wanted to make something that felt very agent-native and felt like a way to collaborate with other teammates.

Yeah.

Um, their dots as well, right? Something that you can kind of curate this set of, of teamwork that, um, you're, you're tagging each other in. You're able to use it in a way that feels like an extension of ChatGPT. And so this is really the entry of Codex and ChatGPT into a multiplayer, more collaborative era. And we're going to do a lot more here.

How do you think about actually, like, the balance between how heavy your hand is on the instantiation of UI on the fly? Like, I noticed this weird thing just happened randomly in Just, just, I believe default ChatGPT. I was asking like, is it a particularly hot summer? And it pulled historical weather data from my local weather station and it created this chart. And when I would scroll over it, it had haptic feedback where the phone would vibrate. And I was like, is that someone at OpenAI like being like, we should put haptics in this thing? Or did the model do that? And then if I love that, am I able to sort of productize that within my organization to be like the way we do charts at this company, there's haptic feedback. Like, how does this flow from what you get for free versus what you put your finger on the scale with versus what an individual user might want as a pattern? That's kind of what I'm—

This is what I am very passionate about, this subject. And I think, you know, a few months ago with Codex, we brought this visualize skill.

Yeah.

That kind of makes UI on the fly. It can either be a full game, right? Yeah. In space, you can make a chess game and play it with other people if they have access to the page.

Sure.

That's kind of crazy.

Finally.

Finally. The people are asking for it.

It's like we're split on this decision. Let's play chess for it.

I think the one that you were describing—

I mean, I like it when I can 1v1 someone in Rust.

Yes.

MW2 out of the box.

Well, and the Dot computer actually has a few games preloaded onto it, which is kind of fun. The Dot can play solitaire. You can watch it if you want to use the remote.

just to play if I just want it to entertain itself and play in the background? I can do that too? Even if I'm not there?

Yeah. I've had Codex play Slay the Spire from start to finish.

I think the one that you're talking about with the weather, that one is an extra challenge because you can't have your instant chat model writing a full web app every time.

Sure.

Right? And so this is one that we've worked on a lot. It's— you know, the research is finally kind of there to make it feel Yeah, yeah.

But then of course, like, like eventually that gets flowed back to the models, the models get faster. I mean, we saw the, the, what is it, 8 times faster is UltraFast?

Yep.

How do you think that that will change workflows for enterprises? Like, I mean, I think there was like a little bit of a groan at the pricing, but Tyler was just like, time is money.

It is.

I want to hit this all day long. How do you think people are responding to like the speed trade-offs? Is that like a new core skill for people in understanding how to size How to size workloads and actually get the work done? Allocate resources? Or is this something that it should be abstracted? It is quickly being abstracted.

I like to say time is money saved both.

Okay, there we go.

Fantastic.

Off the cuff.

It's a great line.

No, I think, you know, ultrafast has been one of those things that for us, like, you just, you can't go back, right? Yeah, there's, it's, you know, It's not exactly the same as intelligence doing something that quickly, but it sure feels like it.

Totally.

Especially when we do these dynamic UI things. You're working in a page with your dot. You tag it in and you say, hey, can you drop interactive chart of our DAO growth over time? And it's there.

It's like you're doing a whiteboard session, but you can do it with all the live data from your business. Exactly.

Right. And it goes off and does the research in your data sources. Writes the little app for you and then you can just say like, hey, share this with my team, right? Yeah. That feels so good with Ultrafast.

Yeah, yeah, yeah, that makes sense. What else are you working on specifically to onboarding? We were just talking about like the boom of solopreneurs. Like who is the core demo that you find the earliest adopter? What industries are are on the cusp of getting online? We were just talking about the, like, the 3D modeling capabilities feels like what we saw in, you know, vibe-coded SaaS products coming to the video game world. But what are you seeing on the horizon in terms of companies and developers here today that might stick out to you as somebody who wasn't here a couple years ago but is now?

Oh, it's hard to say. I mean, I feel like it's just really taken off. I don't know if I don't know if you've seen the chart from OpenAI internally too, which is just basically, you know, engineering was here, research was here, and every other one is like quickly following in sync.

Yeah.

I think there's just some of the things that you do together, like pulling data, you know, researching things in Slack, planning events together. I mean, this entire Dev Day is planned using coding agents, right?

Yeah, of course, of course.

They don't look like coding agents anymore, but under the hood they are. Yeah. Really just a lot of everything.

Yeah. How many how many weeks will you be out of a lock-in until the next six-week lock-ins?

I think yeah.

You're going right back.

That's a great question.

Just repin the tweet.

I think tomorrow.

Okay, tomorrow.

I think tomorrow I'm gonna do another lock-in for six weeks and we'll see what's at the end of that.

Well, that's fantastic.

I can't wait.

Anything else, Jordy?

No, this is great.

Thank you so much.

Come on.

Thank you both.

Crazy. This is fantastic.

Walking around right now.

Yeah, it really is.

Luck out there.

All right. Thanks, guys.

We'll talk to you soon. Thank you so much. Up next, we have Peter Steinberger returning to TBPn, founder of OpenClaw. Welcome back to the show. It's been too long. Last time we talked—

Not really much has happened since then either.

Other way.

Flip it around.

There you go.

There you go.

We're live.

Yeah.

How's life?

Have you ever been on a roller coaster that just doesn't stop? Like, I feel like at some point I see the ups and downs. I'm like, yay! And it's still exciting.

Yeah. Is the roller coaster more wild online with the chattering class or internally with what you're working on?

I think I built up a little bit of a— how do you call it?

A shell?

Yeah. That's good.

There's always too much shit online and some of it is amazing. Some of it is just like, oh, you just, you know, so little.

Yeah, totally, totally.

But I cannot stay away from it.

It's just too addictive. Yeah.

So it's definitely a new skill set.

Yeah.

Like having a—

surviving online.

Yeah.

I'm sure you don't know nothing about that.

Yeah, it's the same thing.

No, but, but, but it's actually insane how angry people get and how emotional people get about like software and just various like software products, right? Um, and you've been on, you've been on the receiving end of like, again, that full roller coaster, right? Where, um, it's, it's pretty surreal. The last time you were on the show was just like you were calling in from, I think, were you still calling in from Europe maybe?

Yeah.

But then I think you almost immediately flew over here and you've probably been here Yeah, it took a little bit.

I had some London detours. We have a cool office there from OpenAI. And now I'm here in this crazy A-Town.

Yeah. I'd like to go back because I think there's so much of your previous work that leads into the announcements today, the moment in AI broadly, personal agents obviously. But can you just retell us the story of Like, what, where did the idea for OpenCLaw come from? Like, what was the initial spark? Like, what got you started there?

My spark is annoyance.

Annoyance?

Yeah.

Like, like I do stuff from— you talk about emotions. I also have emotions. I do something on my computer. It doesn't work as fast or as well as I would. I get annoyed.

Yeah.

But now in this day and age, it's like, you don't have— you can use that annoyance.

Yeah.

And just like, let's fix it.

Yeah.

And then the fact that I couldn't really just talk to my computer.

Yeah.

That annoyed me. I was like, this is so— this all feels so crafty. Isn't there a better way? And then I just explored those better ways. And I keep exploring those better ways. You know, I talked about it inside. Like, my whole thesis is that whatever workflow we think we have right now, it's already outdated. We cannot catch up with the models. You just think about like how much new things you can do because now Luna Like too cheap to matter. Now you have ultra fast. It's like, wait, so I don't have to do 50 terminal windows because I can actually stay in focus? Oh, it completely changes things again, you know? And it's like, we can't keep up building.

Yeah. What do you think, as you look back on the boom of OpenCLaw, what do you think the key ingredients were? Was it the connectors? Was it the connections? Was it being able to communicate to your computer from anywhere?

I think I just showed people a different way.

Yeah.

I use AI? And so many were stuck in this world, oh, it can summarize my email. No, it can install stuff on your computer or buy things for you or do whatever you want that you can do on your computer if you just only give it the right tools.

Yeah.

And I packaged it in a way that was weird and interesting and also felt a little bit different. Instead of giving you a wall of text, it felt felt a little bit more like lobsterish, humanish. But like, it was interesting. It showed people a new way and it had some unique ideas. Like, what about if your AI could actually just be proactive?

Yeah.

You know? And yeah, even though like some of it was like token burning, but some of it was genuinely cool and surprised people. So I think this day and age, it's all about ideas and imagination.

Yeah. Where do you think—

It was such a, I mean, it was such an amazing moment because it, it, it felt like the, the sort of hacker culture that, that, that this entire town was built on, like exploded again in this awesome way.

Yeah.

People getting back into hardware.

The main thing is like, you know, you, you kicked off an entirely new paradigm. You set like the whole industry on fire in many ways. Everyone, everyone started using it, seeing it, and then immediately thinking, how do we make this like commercially viable to be able to scale from hackers that are buying Mac Minis into something that everyday people can use. And I feel like in some ways it's surprising that it, given how good the models are, that it took another 6 months to actually get to this moment right now where there's more excitement than ever around personal agents. Like in some ways, like you saw this, the roller coaster, right? You saw this curve of excitement and then, hey, this maybe isn't ready for like a billion people to use. And then now we're back to the point of like, okay.

Isn't that the classical Gartner hype cycle?

Yeah.

And I honestly, I wish it would have been less hype.

Sure.

It was, it was, it's kind of like this roller coaster where like, yeah, and then like it gets even worse, like holy fuck, you know? So, so I definitely, we had like a massive hype cycle with like inflated things what Open Club would do for you.

Oh yeah.

beside of what it could actually do.

Yeah.

And then you had all these insane stories. And then of course people tried it, like, no, it doesn't quite work like this.

Yeah.

Because we still— there's still limitations. Models were not as good, the software was not as good. Of course, everything's still hard. Everything is still hard. And people were expecting this perfect magical thing. That one guy slopped in the basement.

The ultimate slop grenade.

Um, no, it was really a slop grenade for you. Threw it over to San Francisco.

Well, you were, yeah, you were one of the first people to come out and just be like, yeah, it's all, it's so vibe coded. And you were unabashed about it. And it's like, that comes with a lot of pluses and a lot of minuses, but like, here is the thing. And you know, it's free. Like you can go and make, you can make your own decisions. You can go rewrite the whole thing yourself, one line at a time. time if you want.

And people did many, many, many times.

Yeah, yeah, exactly. Uh, how do you think the— like, OpenAI and ChatGPT is in an interesting position because you have, uh, ChatGPT Threads, you have, uh, ChatGPT Work in the cloud, Codex Cloud, Codex on your desktop. There's like a few different ways to interact with this, and I'm wondering if you have a vision for like what the next year looks like with like how much people are getting into devices, their own hardware, versus Versus things like the cloud just eating more and more things. Like, we went through the boom of the open web, then everything went to social networks and Instagram profiles. And I'm wondering if you think there's like a pendulum that's swinging, or if you have any sort of mental model for where the early adopter or the early consumer will be in a year.

I mean, the current situation with like chat and work, of course it's a compromise. And I learned this the hard way. It is very hard to take people from where they are and introduce them into something new.

Sure.

Without annoying these people or confusing these people. So sometimes you have to make compromises. Is this going to be the final form?

Hell no.

There's people working on something that's way cleaner. What I don't fully know yet is if we will be in a future that is mostly your agent or if it's gonna be whole army. I see— I'm seeing both approaches. I'm more in the, in the, in the first team, but I can also totally see like the second team. And then how all these are like how your personal and your team agents interact. There's still a lot of questions that I don't think the industry has fully figured out.

Yeah, and I, yeah, I always, I always comp it back to Like how human organizations have worked, right? So like you have a big company with thousands of people, you might task one person, hey, we have this like general goal, you figure out how to exactly define it and what it should look like. But then hundreds of people might work on it, but then you don't have hundreds of people report back all back.

Kind of replicate those hierarchies and then maybe your agent, oh, this is something really complicated, let me spawn an agent that manages this and then it manages These little sub-agents. There's so many ideas. I am playing with a lot of them. What really excites me these days is how do we build software in the future and how do we work better with agents? So what I just had so much fun with the last few months is moving all of my team to one place and everybody was freaking out.

What do you mean?

We were so embarrassed on how everyone is talking to their agents. Felt so private.

Oh, interesting.

And they're like, I got pushback. And then like, we did it anyhow. And then everybody felt embarrassed for a day. And now everybody's learning so much. It's like, because we all have a mental model on how we use agents. And it is very different because that was not something we shared. Are you sharing your codec sessions? I'm sure you have some really interesting questions.

Yeah.

And I've actually wanted that, uh, back in the deep research era, I would see Tyler Cowen talk about how he was getting a bunch of value for economic research. And I was like, I would love for him just to be able to click on it, share, share the prompt and the result, because I might never think to ask that question. And we've never really developed a pattern around, oh yeah, it's totally normal to just go on Twitter and share a link to a ChatGPT transcript. Occasionally you'll see people do Grok responses.

They like to take credit.

Yeah, they like to take credit for it. It's like, it's good content. You did the research, or it did the research, you thought of the question.

Where are you going through, you know, the first kind of era with OpenClaw and then now this personal agent boom, where are you actually, where have you seen a bunch of agent, like personal agent activity and you're actually like, I don't think that that will be a paradigm. So for example, like shopping is a good example. Like a lot of shopping with agents right now is like walking up to the front of a store and asking a person outside of the store, like, hey, I think I want to get this thing, and then they— can you give me some information on it? And then maybe they go in and they come back, and it's this sort of like strange dynamic where you maybe at some points it just makes sense to go into the store and see all the information as a— um, and, and so there's like every single one of these, you know, big debate around restaurant reservations in the last, you know, 48 hours as well, right?

Yeah.

So it's like, where are these areas where you've seen some activity and you're like, it doesn't even make sense to really focus on that from a product standpoint?

There's so many companies that want to own the experience. And I don't want to talk to your agent. I want my agent to be able to use your system in the way it thinks it's best. But then they push their own agent and they block my agent and I just go somewhere else.

Yeah.

I, well, ultimately we don't see how that will play out. about, let me, let me use my agent. Yeah, I like my agent. Make your service good so my agent can use it. Well, now we have like the agent wars, I feel. Yeah, like there's big companies that, that block their website.

The agent wars.

Yeah, fucking agent wars.

Yeah, and then here's an example. So shopping, going back to shopping, there's all these, let's say, like companies that make, uh, sneakers, right, that are super desirable, and there's more demand than supply. Well, they've had bots trying to block other bots forever. And now people are just going and they want a generic pair of sneakers and they're getting blocked. Right. And even like an agent that is sufficiently aligned to the user is going to, you can just set it, hey, when this thing drops, go and make sure that we're the first person to buy it. And yet that's a thing that like the companies have been trying to block for—

Well, if you cling on to your old ways without rethinking how else we could do it. Maybe you have to rethink some systems. Maybe the who clicks fast enough is no longer the right way. I don't know. I bought like the Steam Machine and there was just like a window of a week where you could register and then they did the lottery. That doesn't matter if my agent clicks in 0.1 seconds or as a human clicking in 1 hour.

Yeah.

Everyone gets the same chance. So, there's a lot of ways where we just need to rethink how we did things in the past because That it no longer makes sense.

Yeah.

Instead, you want to block my agent? Oh, then I buy something else.

Yeah. How do you think about agents going outbound, becoming more proactive? Everything from OpenClaw to ChatGPT has been very user-prompted. I come to ChatGPT, I ask a question, otherwise it doesn't do anything. It just sits there.

Really?

But it feels like we're about to enter a more proactive— I'm already seeing it in ChatGPT, it'll recommend some prompts. You can imagine the future where I'm more in the response category.

I want an agent that's always thinking. It should be thinking about this interview right now. And then maybe I have like a hidden screen.

Oh yeah, yeah, yeah. Maybe we'll meet proxies. Yeah, you're going to meet proxies.

It is also really hard to do.

Yeah, it's a design problem, right?

Yeah, not just, it's also a hardware problem.

Sure.

Give us more chips.

Yeah, incredible amount of difference.

You want—

Looking at everything on your calendar and saying, you have an interview. So I went and did deep research. Here's what John likes to talk about. Here's what Jordi talks about. And they did all this. They put it together, put it in a video and audio, and they gave it to you in every single way. It's a game.

And I want Astro 6 ultra fast thinking about my day.

Yeah.

And not just looping.

Yeah.

You know, like if everyone would do that, it would just not work today.

This is the country of geniuses and the Roomba. So to decide where to go next, it is—

Down the stairs.

Yep, exactly. It's deploying an incredible amount of incredible amount of superintelligence to solve the most mundane things. But it does feel like that is maybe the natural flow, which is, in the shopping analogy, like the credit card permeated shopping, and it took a pretty complicated dance of, are you writing a check? Who are you? Write the check. Pay with cash. I give you the change, et cetera, to just tap the card. And there's another world where the inventory and sizing is checked behind the scenes All through my agent talking to your agent. And I think any, any worldview that assumes that only one side of the interaction will be agent-mediated is mistaken. I think that, uh, if there might be a world where in the future our agents have talked to each other for the equivalent of a million years to prep us, but I don't think there's any world where, where my agent is spamming you with tons and tons of emails and you're, and you're flooded and you don't have anything, you don't have any defense. The agent wars.

We're going to fight AI with AI.

Exactly. The agent war will happen below the fold, and we will both be blissfully ignorant of the trench warfare that's going on.

I mean, if anything, it keeps being interesting.

Yes.

Nothing's slowing down.

Yes.

I kind of dig that we finally arrived in this world where so many people can now try agents.

Yeah.

But also, You know, there's too many people that just— we build all these super intelligent things, but we're not really good at explaining people what you can actually do with it.

No, totally.

There's still a lot of work to do.

Yeah. What are you tinkering with these days? I feel like a lot of people are getting the hardware bug, the robotics bug, the video game bug, 3D rendering. I went back into Blender for the first time in years. Are there any new— like, I mean, I imagine everything blends together, But are there any areas that have interested you because of the model advances? You're like, oh, I want to go check that out, build a robotic arm that paints or build a video game or something like that.

Robotics is definitely something I bought some stuff and I'll play more. But I'm really still very interested in, as the models get better, how much more can we move up the stack?

Okay.

So you said before about is agents building hierarchies. How can I build this? I want to build an agent that creates those hierarchies almost on demand.

Sure.

If I tell the agent, hey, rewrite Postgres in Rust, I don't know. Figure out, it's like, oh, is Peter okay? But maybe also, okay, he's serious. How can we plan this? What structure does it need to be created on?

And also over what timeline? Because maybe you want Postgres rewritten in Rust and you want to use it. Maybe you want it now. But maybe you want it— of course you want it now. But maybe you want it in a couple of months. And so it's like, okay, well, you're going to throw this on the back burner, cook on it for a while. I mean, I've noticed this where there's certain things where I'm like, oh, this is actually something I'd like to play with tomorrow. I'll set a goal and run it overnight. And if it's rendering something and it's sitting there, it's fine. And then other times it's fast mode. Let's get it done.

Like, I was very unapologetic on trusting the models.

Yeah.

And where people, to the point where people were mocking me, oh, this will never work.

It's slop. Yeah.

Look where it got me.

Yeah.

And yes, it's slop, but also yes, we're going to make it better and the models will keep improving. And then slowly people are waking up and we moved from this world where Of course you look at every single line of code. Like, yeah, we scroll through it.

Yeah.

So, and it's not stopping. It's not stopping. There's still, and there's so much of how we do all those things around this that need rethinking.

Yeah.

And I, I just find that super interesting.

How do you think about your personal workflow? Sam earlier was saying historically you would want to think of 30 ideas and maybe whittle it down to one and just work on that for a few months. It feels like the success of OpenClaw came because you shipped 30 things in 60 days. But are you still trying to do that internally where you're just building, again, silly little kind of projects just to see what's possible to almost generate more robust ideas that you can then invest more time in?

Well, these days I kind of get annoyed if I can't change software with a prompt. I got so used to like having all the tools myself that now I'm kind of annoyed at the operating system, you know, because it feels like it's the last frontier. That's why, like, I started shifting more towards, hey, Linux is now actually cool.

Yes.

Because yeah, there's like a lot of problems, but I like problems.

Yeah.

And, and it's all text-based problems.

And my agent just goes down and reads the kernel and compiles the kernel.

Yep.

And now my webcam works.

Yep.

Yeah, you know, and then that gets me excited.

Yeah.

Like, oh wow, I can now literally own the stack and there's nothing we cannot fix anymore. Like, everything's fixable. Everything is still hard, but nothing's impossible. Yeah, I could write my own OS. Maybe it's still hard, but let's see. With enough tokens and advances, that ridiculous thing will not be that ridiculous anymore in the future.

Well, yeah, it's interesting too because the agent actually becomes the operating system in many ways. I grew up using Mac computers and never had a Windows PC at home. ever in my whole life.

Wow.

Loser.

Jealous. Very blessed. Very blessed. But I got a sim racing rig, right? And so I have a PC for the first time. I'm like, I don't know how to do anything on this thing. But you just set up Codex with the computer you use, and you just talk to it, and you tell it what you want to do, and it just does it.

I built a little ESP device, and it has display issues. So I glued my webcam to the display, and then Codex ripped overnight. And then I told it to fix the display driver.

That's so crazy.

You know, all you need to do is think about ways how your agent can do the best work.

That's so fascinating.

Close the loop. And then the problem was I also told it to fix audio and then I was waking up, hello ESP, hello ESP, the whole night. Because it took so long for it to figure out that the volume was down on the other thing. Well, eventually it learned.

That's so funny.

That's so funny.

Yeah.

Yeah, that's amazing. Are you, you said you put the team together, That was digitally, correct? You said you put all the— the whole team is now sharing what they're prompting their agents, but physically, what's your day-to-day look like? How much time are you in San Francisco?

Oh, I'm here.

I'm here.

Okay.

And I, yeah, maybe basically have 2 teams. I do a lot of work on the foundation side.

Sure.

We push that forward. And then there's a lot of things inside OpenAI where we try new ideas. There's a lot of of, hey, we tried this, how can we make this into something that's a 1P product?

Yeah.

So there's some cross-pollination. I take some ideas back as well because I like to play with them. Some of you disagree on ideas. Not everything works in that scale to that scale. It's a very different way how you need to build things.

Yeah.

But it's kind of cool. So yeah, I have a lot of jobs.

I love it.

Well, thank you so much. Amazing to catch up with you. For coming on the show.

Thank you. What a year.

Yeah, thanks for having me.

Great to see you.

See you next year, hopefully.

Hopefully sooner than that. We'd love to chat with you again. Uh, up next we have Thibault, the CEO of the Reset Company of California.

San Francisco.

Yes, uh, the, the master of resets coming on the show. Thibault, here we go, taking a photo. Can we pan over to them?

I don't think so.

Thibault, thank you so much for coming on the show.

How are you doing?

Man of the hour.

What's happening?

We're going to have you throw that headset on. The microphone goes on the left. I think you got it right. There we go.

I wish you brought the button.

Oh, he already pushed it today.

It's been used.

It was a one-time use.

Single use.

Yes. But the physical button is a great manifestation of all the hard work. How is Dev Day going? What are your reactions?

The vibes are awesome in person.

Yeah.

I feel like everyone I talked to here seems super excited. Maybe not about the things that I thought they would be most excited about, mostly ecosystem and, you know, us opening up like sign in with ChatGPT.

Yeah, we can talk. That was, that was the thing that stood up. I, you know, we, we had it on a screen and I stood up and just started clapping.

You actually did.

Like, you know, I'm watching my team play, you know. But, but to me, that was a standout thing because even, even having Peter on just now, Like, Peter didn't have the ability to offer massive amounts of compute to the world, and so he made a bunch of product decisions because of that, right? The, you know, local approach, things like that. And now if you're a developer and you have product experiences that you want to deliver to customers, you can do that in a way that will be much, much more cost efficient.

Yeah, so maybe reset on that experience. Explain what actually was announced today, and then we'll go into the discussion.

Sure.

But just briefly on that Peter part, I think like the great thing is you can now focus on like building great product.

Yeah.

You know, instead of like focusing like on how do I make the token economics make sense?

Sure.

Yeah, a lot of things were announced, like maybe even like too many things.

Yeah, no, that was my, I mean, I always default when I'm talking to, you know, 70-some portfolio companies. When they're like, we're going to launch this fundraise and this thing all on the same day. I'm like, just break it up, launch it, you know, spread it out over, over 2 weeks. But in some ways, seeing this live, I was like, okay, it actually makes sense. It's fun to have this moment where all these things, and it creates this like pressure around the team to deliver.

Yeah.

So the thing I'm most excited about, because it has changed the way I work so much, and I can't wait to see what people do with it, is like obviously DOTS. Where I think this is just a new way of packaging a lot of the things that we were seeing working before, like very long-running agentic tasks. The configuration that we have, like the agent never stops. Occasionally it sleeps.

Yeah.

But it's always online, always doing proactive research, thinking about your goals, thinking about your preferences. And then you just steer it live. Just, you know, you would give feedback to anyone else. You just give your dog a little bit of feedback. In fact, it learns. And so the more you interact with it and the more you teach it about your preferences, the better a job it does. And at some point, you know, a couple days in, you're like, wow, you know, just like, I can't really quite imagine doing work without this thing.

Yeah.

So it becomes like super delightful, but it's something that you need to invest a little bit of time in.

Do you think developers are already— do you think developers are more in a mindset to adopt DOTS or more excited about sort of diffusing AI throughout their organization to parts of the organization that may have not be, like, not, they haven't onboarded workflows onto Codex necessarily because it's been like intimidating. Is that part of the next leg up here?

Yeah, I think democratizing and they're figuring out like what is the natural way to interact with these things. You know, if you're like, oh yeah, you know, like fire up Codex in the terminal.

It's like, you know, it's capable of editing documents and it's capable of creating slides and you're like, Yeah, and there were some incredible marketers and finance people that were doing that, but there is value to wrapping things up nicely, right?

Yeah, and then also making sure that it's safe and it's secure, like right out of the box. That configuration is really minimal. It does something useful within the first 5 minutes. And you don't have to be super technical, but just kind of still feel the power of it, right? And that's what we're doing with DALL-E. And then in the future, you know, we'll ship a lot of this, the same technology and the capabilities that, you know, read in ChatGPT as well. Yeah, for, you know, our 1.2 billion users. But really, it's about like getting to the essence of like, you know, how much can you simplify really?

Yeah, talk about the current state of the difference between the model and the harness. It feels like this year we went through the whole narrative of like the harness really matters, and I'm interested in the lessons from both the Harness Matters and specifically Codex as a harness in DOTS? How much translated, how much similar lineage is there, or is it possible we're even in a different regime and I'm using the wrong metaphor?

It's like this interesting tension between the model and the harness, and you always build the harness a little bit ahead of the capabilities of the model in order to be like a kind of make up for it. And then the next generation of models kind of catches up to that. And then you have to delete part of your harness because it's actually holding it back. And so this, like, it's very playful and dynamic. It's a dance. Yeah, it's a dance, right? It's this beautiful dance of the model, the harness. The Codex harness is used to power DOTS. So like the same open-source harness that we have. And then we also use it to power the Agents API. We have, which is like, you know, managed agents. And a lot of effort is going into this harness to just make it very efficient, safe, secure. And so I would say like right now, that's most of the benefits that you get from this harness. It's like just really a hardened thing. And then on the complexity of the tools, it's like, you know, we're like just really minimizing that because the current generation of models is just like so, so good.

And is that how I I should think about the specialist dots as sort of fine-tuned harnesses for certain workflows? What is the mental model to sort of deploy those specialist dots across an organization?

We have a lot of specialist dots at OpenAI. And the difference is that you use the existing systems within your company in order to provision the identity and then the access and the machine. So we have them running on Mac Minis.

Okay.

So we provision them like exactly like employees, like we mint like a new little identity. And then, you know, it's like it's got this IAM and the roles. And so we have like, you know, very good security story there. And then we give them like a very premium machine as well because the model is incredibly capable under the hood. Like, you know, DOTS are like beyond PhD level, right? And, you know, if you look at Astra, it's just like incredibly capable at like computer use.

Yep.

So your bot can just like use a computer just like you would. And so by giving them like premium hardware as well, it's like, you know, it just even increases their capabilities further. And so we have these set up and inside of OpenAI to do all sorts of things, including procurement, for example.

Sure. Yeah. On computer use, can you help me understand the levers that are being pulled to speed up computer use? I noticed a jump between 5.5 and Astra, but I wasn't exactly sure whether the speedup was coming from like a fast mode or, or the actual model or the harness. Like, what are the levers? Like, what scaling law are we following to— when I, I mean, I was demoing with a video game like Slay the Spire just because it's visual and I can watch it work and I know how fast I can click. Uh, and I'm interested to know, like, when actually— we keep joking about it, like playing Call of And we're like, well, it's clearly not there yet, but are we— is it a 10x speed increase every year? Like, how do you even— like, what are the levels?

Are we going to accidentally create an aimbot for every single video game ever created?

Maybe. But how much of that is happening in the harness and the model both? Like, how do you even think about understanding the progress in computer use speed?

I do think we have seen a 10x over one year speed increase. And it's not that we made the model sample faster.

Okay.

Obviously fast and ultra fast plays a role if you want to go even faster, but that is not what we improved in terms of computer use. It's both the harness and the model. So when you think about the model, it's like, how much does it need to think before it can take the next action reliably? That's one element that we really improved. The other one is that we take safety super seriously. So we don't try to just click around and let the primary agent do things. Like we have a second agent, which we call auto-review, or internally we call it Guardian, which watches over this primary agent. It's like, actually, oh, wait a second, you're entering, you know, this sensitive information on this website. Like this domain is not quite right. You know, you should stop that action. And so it intervenes there. So we also made this auto-review model faster.

Yeah.

And then the harness, because we had Astra, Tejal talked about this, like we were able to kind of use Astra at scale in order to improve the performance of the harness. harness, do more things in parallel, you know, remove little stalls, and then just really innovate there as well. And it's like, it's all open source, so, you know, you can go and dig and figure out, you know, exactly which techniques we used. And then combined all of this together, you know, then we got this 10x speedup.

Yeah.

Next year is the 10-year anniversary of the first OpenAI Dota competition. I think it was 1-on-1, and then they did the Dota 5 a couple years later. And I, and My big question and prediction is like, will a stock OpenAI model with computer use be able to beat that original model? And it's going to be close because it's got to be actions per minute. It's not just the reasoning. I think if you were able to pause the game and cook on a really advanced model, you'd be able to get there. But the computer use is like the final thing. And I think that's going to be pretty magical.

Yeah. We had ultrafast today on Astra. It's 300 tokens per second.

Okay. So, and maybe I'm going to go demo it today.

Not quite that, you know, one token is one action.

Yeah.

Right. It's more going to be like you need 20 tokens to take one action.

Right.

So, you know, you're going to have, you know, 30 actions per second, which is a decent amount. And then it's all like, how smart is the model? Like, you know, can it actually reason about long-term strategy and movement and all the past actions? And what we're seeing is that currently we announced as well Decisions API. We further tuned the constraints so that you can make decisions extremely fast. So you can make one decision roughly every 200 milliseconds.

Yeah.

So speed is just like, we're just progressing way faster than even I had anticipated 6 months ago.

Yeah.

And I think we'll see, like, you know, definitely, like, real-time, um, real-time actions, uh, within the next year and, like, beating that original model.

Yeah.

So, so, so, uh, yeah, I mean, I, I love those announcements. They're always really interesting. Uh, so many different models and, and speeds and cost trade-offs and Pareto charts and all sorts of stuff. At the same time, uh, how much do you want users to actually Understand that, internalize that, become experts there versus just wait a little bit and it'll be entirely delegated to the harness or to the company even or the product. I'm just wondering how long we stay in this regime of understanding the model picker because there's some real value that comes from being able to, you know, think, okay, we need this, we need this chart, but we need it right now. So I'm I'm gonna, you know, but that could just be in the prompt. Like, go to your dot and say, hey, I need, I need a best estimate of revenue for my business right now. Or, hey, look, we— I need you to go reconcile the books, so get it to me by Monday.

Yes, the way that you think about it feels exactly right. Okay, um, humans don't have a knob, right? So you just, you just tell it in natural language. Yeah, you're like, I need this right now, or I need this tomorrow, or you have $100 in order to achieve this goal.

Sure, yeah.

I'm throwing a big party. I want to spend $50.

It's like, good luck.

But the model should just adapt and the system should adapt and configure exactly how many resources it should use in order to achieve your goal. And that's what we're working towards. With DOTS, in a way, we already simplified things massively. There's no model picker. Your DOTS is just there. It's available 24/7. You can text it. It modulates its effort Depending on the ask. And it might not be perfect. This is also why we're announcing it at Dev Day. We want to build it with the community and like we're learning with the pro users. But eventually it will be so good that you will kind of be shocked that we had a model picker ever.

Yeah. How are you spending, how are you allocating your time? Because I imagine you're sleeping like 4 hours a night. The other 20 hours.

Yeah.

The other, yeah, whatever the other 20 hours a day. How much time are you spending on like recruiting and like who are you actually looking for there? You, you, the team is firing on all cylinders. You want to make sure when you're adding people, they're, they're, they're, you know, speeding up the team versus slowing you down, but recruiting versus everything else on your plate.

Yeah, I spend a decent amount of recruiting, but I try to do more than one thing at once, right? So for example, I might talk to an exciting candidate and have a really fun conversation about an idea that I think either the team is already pursuing or something new and speculative that we're thinking about. It's like I'm kind of trying to see what this other person is also, how they would kind of handle it if they were on the team. And so it's not that I'm actually kind of wasting my time in an interview and doing anything boring there. I'm actively thinking about the work and already simulating them being part of the team, which then also is interesting to myself, like, you know, as it sort of evolves my thinking, sometimes in, you know, quite novel ways. So I would say like interacting with other people and like, you know, I'm always looking for great new candidates to join the team. I would say it's like, you know, 20% of my time.

How do you think about the longer-term ramifications of memory and going deeper with a particular agent and system? Because there's this interesting thing where right now OpenAI has business relationships and there are some companies that have gone all in. And at the same time, there are individuals who have gone all in on Codex and have complex workflows and there's more memory there. But there is a world that I could imagine where if somebody shows up on my team and they're doing their work and I notice that they're searching the internet using Bing, I'm probably not going to say, hey, look, this is a Google organization. But I would say, okay, we use Gmail, not Outlook. You're on this particular email system. And there's some sort of interaction there that feels like, are we going to be more in the regime of you have your agent, your tooling, and I'm hiring you to work with me and you bring all of whatever tools you use? Or is there going to be a regime where, hey, you're joining a particular shop, like a Rails shop, you're writing Rails, you're writing Ruby, uh, or this is a Python shop, so I need to get up to speed on that. Have you thought about where this goes in the longer term?

Yeah, it also, again, it feels like the way that you're thinking about it feels very natural. So just like humans, you know, you don't want to just like re-explain the same thing over and over again. And, you know, if you're a Rails shop, like, you know, you hire the person and you're like, hey, you know, you understand we're a Rails Here's the documentation. Learn how we do things over here and then try your best to do it. If you don't get it quite right, I'll provide you feedback.

Yeah.

And so this is very much the same that we're thinking about it for DOTS and agents in general. Space is like a shared memory system that you maintain across the organization. You have your own notes, but you also have collective notes. And so for example, you could just write in there, like, hey, you know, we're a rail shop. Yeah, like, here are the things that you need to know about. And then, you know, the first thing that it does is just like just goes and reads about it. And then memory system is something that we have quite perfected and are continuing to innovate on, like with the ability to stay, you know, have coherent memory over very, very long periods of time. So you don't want to give feedback today and then learn that, you know, 2 weeks later it's just like, you know, kind of like completely forgotten about it.

Yeah.

And it's just like, you know, painting the House blue again. Like, I really don't like that color.

Yeah, yeah, uh, yeah, it's a fascinating interaction of all these things. But thank you so much for coming, breaking it down.

Yeah, I wish we had more time. I, I think that, uh, the, the intensity around your X account is something that, like, I think very, very, very few, uh, people at companies at any point in modern history have had to deal with. Like, I, I But as I've— as we've gone through this year, just like seeing how much, just how much attention is focused on the next thing that you post and then what you have to go through to actually balance all of that feedback in real time and understand where am I right, where were we off, and then adapt. It's, uh, it's been incredible to watch. So yeah, uh, good luck out there.

Congratulations. Yeah, that's good.

Yeah, we'll see. Maybe the dot's been running the account.

Yeah, that's the big reveal. Thank you so much for coming on the show. Up next, we have Cath Corovac coming onto the show. Cath, welcome to TBPn. Thank you so much for taking the time.

Yeah.

It's a little windy.

A little windy.

Do whatever you need to to get adjusted. Take a seat.

Great to have you.

Great to have you.

Thank you for having me.

Big day.

Yes.

I didn't expect a hug from, from the man.

Oh yeah?

Yeah.

For those who don't know you, please introduce yourself. Tell us a little about your role at OpenAI.

Yeah, I'm Cath. I'm a product manager. I work on GPT Sites, ChatGPT Sites.

Yeah.

Yeah.

What has the traction been with Sites?

Oh, it's been huge.

Yeah.

So let's see. I don't know what I'm approved to say, but I think I can say—

I think I have a doc here. I want the number.

Whatever went out.

10 million sites.

10 million. Okay.

Written in under 3 months since launch.

Yeah. So We launched 3 months ago.

I'm responsible for like 45 of those. I make them for everything. I love them.

Oh, really?

Like my favorite product. I love sites.

Yeah.

Okay, cool. The wildest use case is like looking at— John was like looking at a bunch of different products online and just instantly building this wild— you know, this was like a couple months ago.

Yeah. No, it was basically— I would take— I was looking for a house. I would take a Zillow listing, send the link to Codex, and say, go and download all the photos, uh, restyle them with images, and then put it together in a site where you can swipe left to right to see if this house went through a remodel, would it look better?

And then styling it in, in a way that—

Yeah, yeah. So you would know, okay, this house is this, this style of architecture from this year. This would be a faithful, uh, you know, evolution of this. It wouldn't just, you know, change the architectural style. And then I'd made so many sites that I told it to make a meta site to like put all them together. Of course you did. I've got a prompt to do this. Oh yeah.

And, and I mean, I have a couple, but I'll send it to you.

Yeah.

And like how to organize it all.

Yeah. No, it's amazing.

Because that's like, that's when you know you've like hit the limit.

Yeah. Yeah.

When you actually have to organize this. So did you, did you tell it what your preference is? Or did you give it like a base?

Yeah, yeah, I gave it—

Typically, John has like very specific tastes. So if he was going to go through a remodel, he wants it to look a specific way.

Yeah.

The image model takes that into account and nails it within the context of that space.

But it was so helpful because a lot of times you'll see a house that went through a remodel and it prices higher, and then a house that's pre-remodel, it's lower price. And you're like, okay, well, if you equalize these and Zillow Zillow has a feature for this that lets you sort of like redecorate a little bit on some houses that list, but not all of them. And so this was a nice way to just style it. And then I could just pass it around to like family members and show them.

It was a great experience. So did it show you— this is more real estate questions than it is—

Okay, let's go real estate. We got a whole mansion section. We can talk cars if you want. We can do whatever you want. This is the show. We don't only talk about AI.

But I'm curious if it could do something, and maybe you weren't like asking it to do this. Yeah, but you know, like when you go through a remodel, your home value either like goes up or goes down based on what you're doing.

Yes.

You're like, I'm gonna add a white kitchen.

Yes.

Or if you add a pool, yes, your home value goes down.

Yes.

When you think like, oh, pool is like fancy.

So it did all of that. It looked at price per square footage for remodel homes in the particular neighborhood, block by block, finding comps, building all the comp sheets, and then figuring out, okay, what is the realistic building cost for specific things, creating itemized list of how you'd spend on an architect versus raw materials. And then would you have the, the correct gap to support that based on like a 10-year holding period, basically? So it did all that.

Truly, truly personal superintelligence. Because 5 years ago, the amount of experts and like, okay, I don't have time to build the website, so I need someone else to like build this thing. And I need to be able to like compare and contrast. Yeah, it would be an end.

Yeah. And my end conclusion was just like, it's just going to speed up the economy because more people will be doing Like the remodels, more people will be buying stuff because they'll have more confidence in these things. Uh, it was, yeah, it was a delightful experience.

This is why we built sites.

Yes.

Okay, so we built it for work-related stuff for sure, but also people use it for personal use cases. I do. I just built a site. I don't, I don't want to have to go— I, I like tinkering with hardware.

Sure.

I want to have like, do you guys know the Skylight? Uh, it's like a family, um, organizer.

Yeah, Calendar.

I don't want to go and buy it. I'm sure it's like a fantastic product.

Sure.

But I don't want to buy it. I want to build it myself.

Okay.

You mean the picture frames that cycle through images?

It specifically links to Calendar.

Yeah, but they have their own software and everything that's like specific for families. So I don't want to do that. So I'm building one with ChatGPT Sites.

Oh, and then Sites is the backend basically.

And we just launched plugins in Sites today. And so I can like hook it up to my calendar.

And then with a prompt, you can iterate on it and say, okay, I want to change this.

Exactly.

We're going with this theme. It's Christmas. Make it a Christmas theme.

Yeah, exactly. And I can like hook it up to my kids' calendar.

Yeah, school.

Yeah, do all that stuff.

I think it's really good for, yeah, on-the-fly UI. Like, uh, I was discussing potential guests for throughout the rest of the year, had, uh, basically Deep Research pull a bunch of potential people that are, you know, one step away from guests we've had on the show. But instead of just giving me a big long list, I was like, just build me a website where I can thumbs up, thumbs down, share with the team. Very easy. How does that interface with Space? How do you see these products linked? Is there going to be another merge? Everything's merging all the time.

Oh, there's always a merge.

There's always a merge and an unmerge. This is the life.

Sites are in Space.

Okay.

And that's very intentional. Sites are very— they're a very collaborative place.

Yeah.

And you can add editors to your site. You can add co-editors, co-authors, etc. And it's a very rich environment to to work in. You can like build a site, you can share it with somebody else. We have a lot of people at OpenAI who are doing like— we have our IT team is using this to manage like where everybody is in the world.

Yeah.

And they're sharing those sites with other people like around the world who are like boots on the ground managing security. And so they want to see like, they want to see like, okay, who's flying in today? Is there an event happening? Is there like— how is the weather? Is there a weather event happening? Like all of that stuff.

Yeah.

And so they'll co-editors, co-authors, everything to their site to manage it. Inside Spaces, what's cool is that you can— you have all of your artifacts right there available for everybody. So you'll see us kind of like cross-referencing stuff, making stuff inside Space more and more interchangeable. And so you'll be able to like inside Pages be able to reference Sites and vice versa, and Slides are launching soon. A lot of people are making slides with Sites. You'll be able to put those into Spaces really easily. So it's—

What's the timeline for this to be available to anyone? Like, is there— if you go far enough in the future, can somebody on a free ChatGPT plan create a website?

Oh yes.

Yeah.

So we are very, very interested in making it available to everybody. We have to figure out like infrastructure, how to price it appropriately, etc. Like, we just launched 3 months ago, so we're still kind of learning, but we are also working with the consumer side.

Ad-supported.

Yeah, ad-supported.

Oh, there you go. Your site will have ads all over it, banner ads for everything you do. That's still free.

You know, I'm big—

I want that.

This space is e-com.

Yeah, okay. So we could do, you know, sure, sure, sure.

There's a lot here that we could— we could—

I want to go back to the Angelfire world. I don't know if you remember that, but like it was free web hosting and you'd have these, uh, these banner ads, crazy display ads all over that you couldn't remove.

I'm hoping we can get away from the ads.

Well, we'll have to disagree on that because we are huge fans of advertising.

And so I love, I love advertising it in like a unique way. People are like, oh, this, this meeting could have been an email. And I'm going to be like, well, you're Text me, John. Could have been a website.

Yeah, yeah, basically there.

We're basically there.

Yeah, yeah, yeah, yeah, totally. Anyway, thank you so much for coming on the show. Congratulations on the launch of the Progress. 10 million, 10 million, 10 billion soon.

10 billion.

There's gonna be more websites. Yeah, per person.

So we'll take that headset off. We'll hand it to our next guest. Ben, thank you so much for hopping on. And I believe we're ready for our next guest. We have none other than Casey Knight. Casey, welcome to the show. Good to see you. How you doing?

Has that been on the table?

This has been on the table.

Yeah, the wind's been trying to take it out.

Yes.

But it survives.

Why don't you introduce this? What is this thing? Oh, you have one.

Funny you should ask.

Funny you should ask.

No, so this is the first chromatic transparent, custom chromatic that we made at Modern Pro.

I love the styling, yeah.

For this event today.

Okay, yeah.

We made, I'm not sure if I'm supposed to say the number, we made 8,000 of them.

Wow.

To make sure every attendee could leave with them.

Fantastic.

Leave with one. Yeah. Um, I will say that breaking news. Yes. This was announced on stage by Sam like an hour and a half ago. There's one currently on eBay for $4,000.

I was going to ask you. They got a trade in the aftermarket.

Yeah.

But it's arbitrage.

A lot of people are like, I'm going to wait to get to flip this.

No. Not some enterprising attendee today who saw an opportunity and just got in there and seized it.

Boom.

But the specific integration, of course, Codex, vibe coding, the idea that you can create a custom game that runs on here, is that the idea?

Yeah. You know, like what we do at ModRetro, the reason why we're focused on retro gaming is about the love and the fun of the game.

Yeah. Mine came with Tetris. It was fantastic. Right?

I rarely take Tetris out of mine.

Yeah, exactly.

It's my anti-Tetris cartridge stays in.

It is catharsis for me.

Yeah. I did buy a crazy cartridge that had like some modern technology. I had like 400 games on there and that was really cool. But you know, that sounds super unofficial. Yeah, it was.

But naughty.

But what this is and what it is today is naughty, naughty.

I don't know if I get in trouble for buying it. I think the person should be sold it, maybe.

We built a custom plugin.

Yeah.

Um, for Codex.

Yeah.

And it enables you to sort of vibe code games.

Yes.

What's special about it is that you're not, you're not constricted to playing that on an emulator on your laptop. You can like, you can flash it to this cartridge immediately.

Yep. It's amazing.

And I think like the best way of framing it is like Torin, our CEO, and his 5-year-old vibe coded a game. And then he was able to flash it and hand the Chromatic to his kiddo and have him playing it right away.

Yeah. And it's something that like leans into the constraints. Like you see, I've seen a bunch of demos that are like, oh, this new model's so powerful. I just gave it one prompt and it made a AAA Call of Duty GTA 6 video game. And I'm like, okay, let's, uh, let's calm down. You made a 30-second video of something that actually looks kind of janky and probably doesn't have anywhere near the breadth of features just yet. We'll get there. But for now, This is the sweet spot. Like, I know that you can actually deliver something that's fun and entertaining and full-featured and not buggy when you're constrained to this screen and just these buttons. And you're not—

also, also the exciting thing is having, having a TAM of one, right? Like, there's so many, there's so many great ideas. Yeah. And when I say that, I'm not saying, okay, the product, but I'm saying like, there are great games where for a 5-year-old, yeah, like, he— they will be the only person that appreciates this game.

Yeah.

And it's combining different parts of their interests into something that would never have commercial appeal, could never get shipped or made.

Totally.

Now it can. And I think that's been sort of the story of every single category of AI use case where you can now make things that, like, again, the example you were giving 10 minutes ago.

Yeah.

It's like you would never spend the money or the time to make this comparison website only for yourself on the fly.

Yes.

And now you can.

Yeah, exactly.

Constraints is a great characterization because I think it dovetails with like what our mission is at ModRetro, which is like embracing those constraints, which amplifies like the why part. You know, like we launched the M64, which is our Nintendo 64 console. We released that a month and a half ago and doing incredibly well. But so much of the feedback we see is from like old people like me.

Yeah.

45. I was 15 when Nintendo 64 came out. Yeah, of course. You got a dad or you got a mom with their little ones, and it forces you because of those constraints, like, physically sit with my kids and play Mario Kart. And when I think of that versus like when we catch our daughter, she's not allowed to play Roblox, but we catch her in her bedroom, like hiding in the closet by herself with her iPad 6 inches from her face playing Roblox. This just feels so antithetical to that because it is about like the social interaction. social aspect. It's about the camaraderie. It's about the fun.

Yeah.

And I think like that's, that's what today in this partnership was emblematic of. It's like you can create something that's fun. They say, check this out.

Yeah. It's so funny that like, I think a lot of— we're all parents here and I think a lot of parents are going through this. Like, I just want the same experience that I had for my kids. And then there's literally a company that's like, yeah, you can do that. Yeah. It's such a simple value prop.

It's, it's a funny, it's a funny philosophical exploration because there's a part of it which is like, is it just our old man nostalgia?

Yeah.

And, you know, Palmer, our founder, has like a really curt response to that, which is no. Yeah, it was just better. Yeah, it was better than— and I think about that because like, you know, we bought our daughter this summer a point-and-shoot camera.

Sure.

And it was like the biggest fucking deal for her.

Yeah.

And like, she has a camera on her iPad, but having a point-and-shoot camera This single-purpose device, it meant so much to her. And I think that we're starting to see like technology start to go full circle.

Yeah.

Like we have these stupid little black rectangles of doom that live in our pockets that you can do everything on.

Yeah.

But there's something about whether it's like a wired headphone, point-and-shoot camera, a device that you have to physically put a cartridge in to play a single game.

Something engaging.

Yeah, I think it's about the relationship between the person and the technology. I feel like we've gotten so far away from that with the rate of innovation getting away from us.

Yeah.

Yeah.

I didn't mean to go so deep.

No, no, no.

I think a lot of us—

no, I mean, I've spent— I mean, every— I'll text John like twice a month, maybe every week, where I'm like, everything that's good hasn't changed in 60 years. Right? And in many ways, like having like machine intelligence is the only thing really that, that I feel like has been super meaningful that's, that's changed in that period. But when it comes to stuff specifically in the physical world, everything that's good has existed for—

Old restaurants, old hotels.

Yeah, cars, cars got worse, you know, a bunch of things, you know, homes got worse, all these different things.

Toilet paper rolls or paper towel holders. Is that your argument?

Oh yeah, that's probably—

He wants a paper towel holder invented in the 1960s.

Yeah, I've got one that has a Wi-Fi connection.

Okay, you're on the opposite end.

It sends me alerts when I get low on toilet paper.

Yeah, no, but the cool thing is it feels like— I really do think that all— I do think all the hardware that we use is gonna like change dramatically in a way over the next 10 years that it didn't in the last 10 years. And part of why I think that is because you have thousands of super smart people working on these new paradigms and form factors, but then you also have the software changing. And the first time we had Peter from OpenClaw on, and he was just saying like he's now getting frustrated with the OS. It's not just the things that are happening as part of the— or on top of the operating system. He's like he wants to go all the way because there's things that can just make more sense when you can generate an operating system almost on the Yeah, I think it's holistic.

I think there's room for everything, but I think what's been caught up in this like tidal wave of innovation that I felt has sort of hit like a really aggressive inflection point in the last 10 to 15 years is that it sort of consumed everything. And I feel like we've kind of reached this plateau where we're able to pause and identify little areas where like, no, no, no, that doesn't need that kind of innovation. I think back to like, if I'm— if I'm— if you'll allow me to be romantic about it.

Please.

Like, I got— I think for my 10th birthday, I got one of those yellow Sony Walkmans. And I fucking treasured this thing. It meant the world to me. It was my most prized possession. I took showers as a kid, but I would take a bath just so I could listen to music. Yeah, in the bathtub. Yeah, I had a Walkman that could get wet. And I think about that now, and it's like, technically Spotify Spotify is better than having Led Zeppelin cassette in my yellow— it's better in every way technically. But it, like, Spotify will never mean to me what that device meant to me. And I think this romanticization of technology is something we've drifted away from.

And the change in medium changed the message. You used to need to listen to a whole album.

Yeah, for me it was iPod shuffle. Oh, which one? The little white one?

The clip one?

The little square one. Like, to me, to me, that was, that was the pinnacle of technology as a kid because I had, I had this thing that was like a quarter. You could like throw it up and down.

Yeah.

And yet it was so powerful because music is so emotional.

Yeah.

And like, it felt like it could take you on a whole journey, be as entertaining as watching a movie, just playing songs.

And you never knew what song was coming next.

Yeah, exactly.

Part of the romance.

Yeah, it was like gambling, gambling with audio. Slot machine every time you press skip.

Yeah, and I think we're gonna see— I feel it now, but this is my world that I live in. I think we're gonna see this sort of looking back. Like, I, I run across the Brooklyn Bridge 7 days a week unless I wake up in San Francisco.

Sure.

And what's interesting that I've noticed in the last 5 years is it used to always be people's iPhones, and now the amount of people I see with these antiquated 10-year-old point-and-shoot cameras And they do it, and it's sort of a flex.

Yeah, of course.

They're using this camera that is objectively a lower quality camera. That's what's on their phone. Yeah. Yet it says something about them.

Totally.

In some ways, New York City is probably getting captured on film more now.

Sure.

Than when film cameras were like the only way to actually take a photo. Yeah. Because the diffusion, like historically, it probably wasn't like, you know, thousands of people spread across the city at all times. times getting every single angle, and yet now that actually is what's happening.

It's constant. Yeah, my favorite 35mm point-and-shoot is this Olympus. Okay, okay, it was $150 when it came out in the, in the aughts. Yeah, and they stopped making it whatever, 15 years ago. Every time I'd break one, I'd just buy it off of eBay for $50. Sure. I noticed a couple years ago that they weren't $50 anymore, they're like $100.

Oh wow.

Okay, then it was like $200. My wife asked me for one for her birthday and I went to buy it, it was $400.

Get a corner of the market.

And now they're almost unavailable. And I think it's akin to what we see with classic cars. People appreciate what these things were.

I mean, that's what makes me so excited about Palmer's. Palmer's teases me. He can't say anything, but Palmer's teases like, I don't like connected TVs. I don't like music. And I don't like cameras. And I'm like, yes.

The, the connection, I mean, I have this thing of like every, the, the frustration that I have with, um, hopefully this isn't just like my issue and I'm like not updating the software or something.

Let us hear it. But like, let us hear it.

But, but the, the level to which, uh, like modern TVs, the software degrades to the point where you're like, okay, like I'm gonna click. Yeah, I'm gonna click. And I click too many times and now I gotta go, you know, like—

Have you followed along with this LG Yeah.

Okay, break it down for those who haven't.

Okay, the TL;DR is that the most modern LG TVs, they have the best panels.

The best panels. Yeah, OLED. It's amazing.

The most expensive. A couple of hackers, a couple, they have a YouTube channel. I apologize, I don't remember the name. They deserve a plug. Yeah, but they broke down all of the software and they determined that the microphone that was intended for like, change it to HBO, it's actually listening all the time. So much so that when it wasn't networked, when it wasn't connected to the network, it would store the audio files locally, and then when it reconnected, it would upload it. I watched that and I was like, fuck, this is what Palmer's going on about with his desire to make a dumb TV. And then I realized in my bedroom where I sleep with my wife most nights, some nights with the guests, you know, marriage, we have that fucking LG TV.

I know, I have one too.

So I unplug it from the wall now when I walk in the room. bedroom. Yeah.

And it's just like, no, but you remember how fast, how snappy old TVs were? It was like you press a button, it was just instant. How snappy? I'm 45.

For me, it was one of the, it was a physical dial with a long wire there. Immediate.

Yeah. Yeah. Immediate. Immediate.

You know, we've got to go back to wired remotes.

Yeah. Yeah.

I'll be the first one to buy that.

I, yeah, I remember, uh, uh, sort of some of my frustrations with the, one of the, one of the Meta Quests was that I found out that I had seven passwords because I had a keypad, a code, a Meta password, Instagram, Facebook password, and an Oculus VR password. And and I and I showed this whole process of like what it takes to just take it and then play a game. And it was like fifteen minutes. And then and then I play and then I just found a video of somebody turning on N64 and you just bing bing and you're You just turn it on. It's the same thing.

You're playing the game.

Yeah, you're playing the game. It's ridiculous.

Our Samsung TV.

Yeah.

In the other— in the guest room.

Yes, yes.

In the doghouse.

Yeah, in the doghouse. In order for me to download Netflix, I have to sign into Samsung. I don't have a Samsung password.

Why would I?

And I don't want one. I just don't watch fucking Netflix on this thing. You are dog-faced, Samsung. It is just so much. They ask so much of us.

I know.

And it's not better for the customer.

No, it's terrible.

They just want to suck that data from you to sell you more ads.

It's terrible. It's terrible. Flipping over to the creator economy, the creator side of things, I have this idea that with the advances in vibe coding, there is an opportunity basically for someone to make a game every single day for Mod Retro or any platform, web games. And that muscle of doing something small every day would compound. I don't know that they would become the next GTA 6, but they would become It becomes something. And I'm wondering how you think about creativity and daily rituals, new platforms in a time of rapidly decreasing cost to create whatever it is.

Yeah, I have a super cynical perspective on it and a wildly optimistic.

Okay.

So the optimistic one is like, what is it, a million monkeys at a million typewriters and they'll eventually type out all of Shakespeare?

Shakespeare, yeah.

I think that's— there's a— there's a— math says that's going to happen. Yeah. In other words, vibe coding games for—

So you were the monkey on YouTube that typed Shakespeare?

Correct.

Yeah. But I didn't have vibe coding. No. And I didn't have AI video. No. And when—

But you did have GoPros and Canon 5D.

iMovie. iMovie and YouTube.

You didn't need to shoot on film.

It didn't matter. I just kept uploading. Yeah. And then like the line that I share that often shocks people is I was, I think, very successful on YouTube. Very successful. And after 7— after 7 years of a very successful presence on YouTube, I had 280,000 subscribers. And then I started the daily vlog. I started— I was the monkey that never left the tree. 18 months later, I had 10 million.

Yeah, it's crazy.

And I think if I had maintained that cadence, if the top of the funnel hadn't become more accessible, the cameras hadn't gotten cheaper, I hadn't gotten faster at editing because the technology enabled that, that inflection point would not have happened.

Sure.

So I'm— yes, I am like the— I am the original monkey at the typewriter. But I think now, especially what AI is doing, it's something that is much more complete. And when people shit on AI video, and when I say people, I mean me.

Yeah.

I think it's gonna get to a point where there's gonna be something that's made on AI, that's 100% AI-generated, that's undeniably good. And it's just gonna sort of force people to be like, I hated it because it was made with AI, but it was a really good fucking movie. Yeah. And eventually that negativity is gonna start to wane. It will be sort of undeniable.

Sure.

There's some cynicism in there. I also think that that will enable, especially when it comes to things like gaming. Yeah. Creating software. Yeah. It's just when you open the top of the funnel that wide, Yeah, the amount of shit being made increases. Yeah, but so does the amount of quality stuff being made. Yeah, so I am highly optimistic that there are gonna be these diamonds among a sea of coal, lumps of coal. Yeah, that, that will have real impact. Yeah, meaningful.

I would also— yeah, when I, when I think back, like, I used to work on little like iPhone games, and I remember when I was like trying to just make Pong, right? It was like I was just trying to make the most simple possible game. The amount of frustration where you go from, like, you know, it's exhilarating because you're making progress, and then you get stuck on something. And I was probably like 11 or something at the time. Then you get stuck on something for like 2 hours, and it's like pure pain. And you're like, how do I figure out how to get through? And then you get through. And part of that is like very— part of that is like extremely positive and worthwhile lessons around hitting blocks and pushing through. But the amazing thing with AI is like just being able to bust through those like much faster and then compound, compound those learnings and, and do an individual person doing 10 times as much in the same period of time. And it's hard not to be excited about that.

It's like a monkey with 10 hands. That's the best analogy for AI.

We have to keep this monkey analogy.

Yes.

Well, on the video side, the thing that I've been, I think naturally everyone went from like, oh, okay, I'm reading way more AI-generated, uh, text. And, and that was over basically the last 2 years. And now it's getting harder and harder to tell. And I, we even know people that produce a lot of AI written content, but they've figured out ways to make it not, you know, pangrammable. Um, at least, uh, obviously. Until you break your silence. And now, and now I've just noticed like way more, way more like AI video, which, which is like working for me. on some niche, like where I'm like trying to learn specifically about a topic. Um, but then, but I keep getting hung up on this idea of like, did the person that published, that has, owns the account, did they actually review this video? Because I think that that's where the, uh, like I'm looking— Yeah, yeah. Is there actual curation? Is there someone saying, because specifically I'm thinking of, uh, There's an account on Instagram that does like sim racing tutorials, lessons, or whatever. And so they're putting out like probably 20 videos a day on different topics. And a lot of it feels like generally right. But where it will break down is like if you just— if the AI is effectively like reading an online forum and then just piping it into video, and then that advice is like expired.

Yeah, you're describing the modern news cycle. Yeah, like that is what, you know, Instagram, TikTok, and, you know, X news is. It's a variant of what you're describing.

Firehose.

That's why there's so much— the noise-to-signal ratio has just gotten so convoluted and difficult to navigate. I don't know, maybe like a more acute example of something like that is like my favorite genre on YouTube is short esoteric World War II documentaries. Sure. There's a guy, he's my favorite YouTuber on the planet, his name is Mark Felton.

Okay, I love that guy.

You know this guy.

That's how I fall asleep to that kind of—

Me too, every day when I get my sandwich at lunch, I sit in my office by myself, I watch one of his stuff. Like a 50-year-old British historian. He is an objectively terrible filmmaker who makes the single best videos. Yeah, it's just him narrating it over whatever images he could find. watch that, and then I also watch these 100% AI-generated that clearly have a human editorial oversight because they're factually correct, very well done, and they have all of the YouTube AI notifications all over them, and it doesn't bother me.

Yeah.

It doesn't bother me because it feels like I'm reading something with visuals that are simply there to support it. It's— I'm very conscious that these are artificially generated images. They're just meant to sort of visualize what something looked like that wasn't actually captured. And for me, like, those are the fissures I'm seeing as someone who's like an anti— who's a purist when it comes to filmmaking. Yet I'm willing to tolerate that. And I think that delta is going to—

That's why I don't— I never— like, AI content is already here. It's happening. There's going to be 10,000 times more of it in probably 2 years. But I haven't been— but I haven't I've been afraid because when I, when I think about my favorite creators and where I'm actually spending attention today, it's like, it's, it's like the guy that you're just talking about, Mark Felton. There's another guy named H.I. Sutton who's just obsessed with like naval warfare.

Yes.

And just nerds out about it. Yeah, yeah, I, I figured. But like, he'll just basically press record and he has a presentation, like slides. Yeah, he's going through it, he's narrating some video. He's like, I don't use AI. just sort of sit here and talk. And it's like incredibly captivating and interesting because you're listening to somebody like the world expert on the topic who maybe uses AI to learn about some stuff now, like off, off air, but then he's just narrating it as a human. And there's just all, I'm always going to have time in my day for the H.I. Suttons of the world.

Yeah. Look, it's a, it's a proliferation of, of podcasts on YouTube. Yeah. Like if I'm watching something When I edit my own videos, like, I will debate a single frame, a 24th of a second. Does this fuck up my timing?

Sure.

Yet then I'll click play and watch a 3-hour podcast where it's just some dude talking for 3 hours because I care about the subject matter. And I think this is like the extraordinary range in people's appetite for what they're willing to give their attention to. And the idea that there isn't room in there for AI content, I think, is misguided at best.

Last question, what shirt are you wearing under that jacket?

Well, I would like to— I do have a plug today. This isn't to sell anything because these aren't for sale, but—

There we go.

Our former intern at Mod Retro, Jeremy, who's now— his title on his business card is Special Operations. Not sure we knew what else to give him as a title when we put him on payroll. He designed this latest Mod Retro merch.

Yeah, so everyone watching can screenshot it and as a fun game, try and identify each of these letters because they each come from a different piece of a game.

Which one got us hung up this morning?

Uh, I think it was the T. We were going— you— I knew the T. Uh, I identified the M, the O, the D. I think I might have missed the E, the R. You knew the O, and I, I, I was off on the 64, But I was directionally correct. So, uh, no spoilers, but plenty of teasers. Go figure it out. Compliments to Jeremy. Head over to Mod Retro, get yourself a Chromatic, and start—

Turn the t-shirt into a playable game.

Yes. Yes.

I'd download it. I'd flash it on my card.

Well, that's our show, folks. Thank you for tuning into TBPN. We will see you tomorrow at 11 AM Pacific sharp, back in the