AI Is Crossing the Frontier of Human Knowledge | 2050 Science by 2030?

You have no excuse if you've got an interesting idea. You can now create

interesting idea. You can now create anything that you can think of. The

models can now solve problems that humans have never solved before. Going

beyond the frontier of human knowledge.

That's how AI, I think, and AGI will really change our lives. Why not try and accelerate science, bring about the science of 2050, but in 2030 instead?

That's Kevin Wheel, vice president for science at OpenAI. After leading product teams at Twitter, Facebook, and Instagram, Kevin is now focused on accelerating scientific discovery with

AI. In this interview, he explains how

AI. In this interview, he explains how AI models are pushing beyond the frontier of human knowledge, why robotic labs and long horizon reasoning could transform research, and what founders

need to understand to build successfully in the age of AI.

All right, this is an incredible time to be alive. I think now you helped build

be alive. I think now you helped build and scale some of the most important technology companies of the last decade.

Uh Facebook, Instagram, and Twitter. And

now you're doing that at OpenAI. I did

ask ChatGBT what it thinks about you. Um

and so uh as you would expect, it was very complimentary, but you're you're also like a very accomplished guy. So

Kevin is thoughtful, low ego, and unusually grounded for someone who's been at the center of so many highstakes products. So how did you like as you

products. So how did you like as you looked at those all those four opportunities plus many others that were amazing? How what gave you the

amazing? How what gave you the confidence that this is the type of company? This is the team I should be

company? This is the team I should be you know working with.

Yeah. Number one advice marry up. Uh it

was my uh my wife originally actually I was a I was in grad school doing a physics degree and I met my now wife who was a Mayfield fellow at Stanford and actually worked at Andre for a little

while. Um, and she was the one that kind

while. Um, and she was the one that kind of opened my eyes to everything, startups in the valley and all of that.

I I grew up in Seattle. My dad was a an engineer at Microsoft for a long time.

So, I grew up programming, but still was just like, you know, math and physics, math and physics as I went through grad school. Uh, and it was my wife

school. Uh, and it was my wife Elizabeth. She introduced me to Twitter

Elizabeth. She introduced me to Twitter back in the day. Um, because she and Jessica Varilli knew each other from Stanford. And after seven years uh at

Stanford. And after seven years uh at Twitter as it grew, she and Kevin Cyester were also Mayfield fellows together at Stanford. So um that's how

that connection happened. And so you know a bunch of these things were just my wife me just following the coattails of my wife. I used to I used to call Sam

wife. I used to I used to call Sam periodically before whenever whenever I would like be thinking about doing something new. Um Sam and I didn't know

something new. Um Sam and I didn't know each other super well, but we knew each other well enough to to, you know, to do a quick phone call and because he he always has his like hands in lots of

different things. He's like doing fusion

different things. He's like doing fusion startups and um and all of this stuff. I

remember talking to him like in 2020 or something and he was like, you know, AI will not replace blue collar jobs first.

It'll replace white collar jobs first.

Coding is going to be one of the big things for AI. This was 2020 and none of us used AI particularly much, at least not outside of like classical ML models that you know, ranking your feed and

stuff. And I just remember being like,

stuff. And I just remember being like, "Yeah, whatever, dude." You know, sure, but like talk, let's talk about something. And so anyways, I um the open

something. And so anyways, I um the open AI thing happened because uh I called him and this time he was like, "Actually, you know what? We have this role open. You should come talk to us."

role open. You should come talk to us."

As soon as I did, I was just like, "I don't like I'll work for free. I don't

just this is the most interesting thing in the world. Um and uh if you give me an offer I'm coming. So fortunately he did.

And then the the original mandate uh as CPO uh with uh with Sam was to to what and what are you sort of most happy with what you accomplished during that time?

Yeah, originally it was uh CPO so leading our consumer products, B2B products, developer products, etc. And um I mean man I think it's we grew like

a weed. I've never seen anything grow

a weed. I've never seen anything grow that quickly in my entire life and I think brought you know kind of brought AI to to a whole bunch of the world. So

very proud of that work. few months ago it was getting clear that our models could not just be great inside chat GPT or inside codeex which I think is an

incredible product but we're at the level that they could start to answer frontier scientific problems like the models can now solve problems that

humans have never solved before so a lot of people like the criticism of AI is oh well it's just it's just bringing together different ideas from different places and summarizing them for you to

give you an answer like I can't actually do novel thinking but we've now seen there have been I don't know what 10 or 12 just in January 10 or 12 open

mathematics problems solved mostly by GPT 5.2 too now a some a few recently by Gemini models are going beyond the frontier of human knowledge and I wouldn't I wouldn't claim yet that they

are solving problems that humans can't I think if you took enough people and you know applied enough mathematicians towards some of these problems they would have figured it out but they had not figured it out yet the model went

beyond what we had ever done as humans and that's pretty cool and that's today right if there's one thing I've learned over the last few years it's that you go very quickly from models could never do

this thing. it is beyond the capability

this thing. it is beyond the capability of AI today to models can just barely do this thing and it kind of sucks at it and like it's wrong most of the time but

you get these glimmers of like ooh they can almost you know do this maybe it only works 5 or 10% of the time and then 6 to 12 months later it's like models are great at this thing and I would

never I would I would always use AI anytime I ever do that again like in eval language you go very quickly from

like zero to five or 10% to like 60 80%.

You know, we are clearly in that middle phase with frontier science and AI where you have all these glimmers of like, wow, it can do something that we never thought AI could do. So, you know, what

more interesting place to apply it than science? I think it may be the most

science? I think it may be the most tangible way that we all feel the impact of AGI. If we dropped like GPT9 inside

of AGI. If we dropped like GPT9 inside of chat GPT for you today, I'm sure it would be awesome. But maybe even more awesome would be that we have all of these new materials and we have superc

conductivity and we understand the nature of the universe and we have personalized medicine and you know like that's how AI I think and AGI will really change our lives. So

and fusion power man like why not why not try and you know accelerate science and bring about the science of 2050 but in 2030 instead and that's our goal

you know when did you have your claude weekend moment or your codeex weekend moment where you're like wait a second the world has fundamentally shifted and everything that I've done before is going to be very different going forward.

Yeah I so I have a very different take on this. I I think it's awesome and

on this. I I think it's awesome and exciting like when you when you take something that has historically been the craft of a relatively small number of people. There aren't that many people in

people. There aren't that many people in the world that know how to program. I

don't know like 30 million people maybe.

And you expand it by a couple orders of magnitude. You get an explosion of

magnitude. You get an explosion of creativity because lots of people have ideas and sometimes they weren't they didn't have any route to actually implement those ideas. Um, and you go everything from like thinking about

people that can start companies now that didn't used to be able to start them all the way through to like I remember sitting and talking with um this was like a little bit postco with uh this

city official somewhere, you know, like small city and he was telling me about the kinds of programs that they wish they had been able to to run and operate just like basic kind of information

awareness things for the people that lived in this small city. It's like I just didn't have any way to do it. The

data was sitting there. I could, you know, draw what I wanted it to look like, but I had no way of getting it done because he wasn't an expert and they didn't have, you know, however many thousands of dollars to to hire a a

person to do it for the city. Now, what

would you do? You would like enter a prompt into Codeex and it would be done.

So, I think these kinds of things are awesome. I'll say personally we just

awesome. I'll say personally we just launched Prism um like uh what a week ago um anybody see that use Prism tried

it uh it's like an AI native environment for scientists to do scientific writing and collaboration so if you're using latte things like that because it's a small product in a small team I've been spending a bunch of time uh like writing

code fixing bugs etc which I haven't done in a bunch of years and it's super fun it also uh I I I went through this transition in the middle of this, you know, because most of my day is meetings and, you

know, we got a bunch of stuff going on.

I went through this transition where I I I was sitting in I don't know, I remember I think it was like Fiji, a meeting with Fiji and I closed my laptop because we're all trying to be better about not multitasking because that's

like one of the worst habits that people at OpenAI have. Everyone multitasks all the time. I had not gotten a codeex job

the time. I had not gotten a codeex job running before I closed my laptop and I was like I just wasted an hour like not because of the meeting but just

because I could have been multitasking during that hour. My my codeex agent could have been like fixing a bug or implementing a feature or doing something for me and now I have to sit

in this meeting just like you know like a farmer being in a meeting. Come on.

And um so and you know the same thing before you go to bed at night you're like okay what like really hard task can I give codeex and just let it chunk away

for like 10 hours. Um so like that to me is a is a different world and if you're really good at it you are not just juggling one job. You've got you know three or four things running in parallel across different work trees and like I

don't know what a cool world.

Yeah. Uh the creativity that's going to be unleashed is just incredible. And I

mean I I think about it as like if you were a furniture handcraft furniture maker like 100 years ago or 150 years ago industrial revolution hits all of a sudden there's factory mass-produced

furniture that's better than yours or at least equivalent right and you're saying wait a second my whole world has changed but now you can design something that can be used by so many more people all at once but like it's this

disorientation that people feel uh because it's so disruptive and so quick at the same time.

I don't know. I think we'll end up with like cuz now you still people still value custommade furniture. I think

we'll end up with like bespoke websites like this was done by a human.

Exactly. Exactly. Same as like you know people will be like I drive cars like as a hobby. Right. So

a hobby. Right. So

um like get off the road you're right. Yeah. Stay over there in like

right. Yeah. Stay over there in like your your little s you know little area for human drivers with things moving so quickly. Like you gave an example of

quickly. Like you gave an example of like how you stay up with things like you're you're you know doing work with your team. You're staying close to your

your team. You're staying close to your team. Like do you have any other ways

team. Like do you have any other ways that you're both able to lead the team, lead the strategy but also stay close to all the developments that are happening within your company within other companies? It feel that feels like more

companies? It feel that feels like more than a full-time job in its own right.

Yeah. I mean the industry is just moving insanely fast, right? I've never seen anything like it. Um, it's exhilarating and it's fun and it's also it's a lot to keep up with. But I think that I I just

think this moment kind of selects for people who are high agency that because you can now create anything that you can think of and you have no excuse if you've got an interesting idea not to

like get, you know, codeex thinking about it while you do something else.

Whatever you were originally going to do in the morning, keep doing that, but have codecs working on your idea in parallel.

Sometimes you'll wake up in the morning, have an idea, and have a thing implemented by the time you're done with the day in addition to doing what you thought you were going to do during the day. So like people that are high

day. So like people that are high agency, people that are really curious, people that learn quickly, those skills are more valuable than ever in this moment. And you know, kind of whatever

moment. And you know, kind of whatever the future holds, I think those skills are going to see us through.

Yep. I've heard you talk about a vision of both the experiment design but also then the experimentation itself and the validation uh that's very you know it was very captivating to me maybe you

could share it share it here a little bit more this open AI for science group is only a few months old although in some sense of course open AI has always cared deeply about science so I I feel

like even though we're a smalish team we kind of have the might of all of openi research at our back because every researcher at openai cares about science And scientific data has been one of the

ways that we uh you know have have improved our models for a long time. But

um so we started thinking about math, physics, theoretical computer science because you can do everything in silicone. You have like closed loop

silicone. You have like closed loop systems that you can optimize and you know is part of this is teaching the models to answer really hard scientific

problems. teaching them to think not for like 10 minutes or maybe the hour that you can get GPT5 Pro to think if you ask it a really hard question, but teaching

models to stay on track for a day, two days, a week, two months at a time to answer even harder problems. Because just like you and me, like if you you could give me problems that I couldn't

solve in 20 minutes, but I could given, you know, two hours. Same is true of the models. the more time they think, the

models. the more time they think, the more the more impressive problems they can solve. So, you start with things

can solve. So, you start with things like math and physics, but then, you know, some of the the biggest ways, the most important ways that that accelerating science is going to feed back into all of our lives in a positive

way is through stuff that we can feel in real life. It's like, you know, one of

real life. It's like, you know, one of our relatives survives cancer because we've made advances in medicine. You

have new devices and materials because we've been able to make advances in material science. And those things

material science. And those things require labs. You can't do those just in

require labs. You can't do those just in silicone. Although I think the the

silicone. Although I think the the importance of simulation is going to go up pretty meaningfully. Um because we'll be able to apply huge amounts of compute to these problems. But then you're still

going to need experimental validation.

You're going to need to try things in the real world. I think it's going to be a while before we have a model that can you know first principles go from like a quark all the way through a model of a cell all the way through human biology.

Like you experiment matters. So, you

know, you start to think about how you do that at scale.

There's lots of opportunity to partner with existing labs and we will but I think there are really interesting you know the science of the future will definitely involve robotic labs and

reinforcement learning loops that go through the real world where the model is thinking maybe running a simulation thinking some more refining the experiment that it can run using the best possible parameters and then

sending that to a bunch of uh robotic labs which by the way you can scale horizontally having the experiments run in real life.

Model the results come back to the model. The model thinks runs more

model. The model thinks runs more simulations thinks and you have this you have like multiple loops. You have tight loops with the model thinking and the simulation you have longer loops you know that go through the real world.

That's how a lot of science is going to be done in the future and you know that is its own form of acceleration. You

think you have you have robotic labs that you can scale horizontally that can run 24 hours a day. You know they're not grad students pipetting things that need to take breaks and sleep and things. So

um and and then the grad students can do things that are much more you know you know leveraging of what makes us human than pipe heading things right

so I'm I'm quite optimistic about where this goes and about our ability as a society to accelerate the pace of science very meaningfully how far away do you think we are from

that what are the key like technologies is it you know robotics that need what needs to happen for us to unlock that fully I mean some of that piece is robotics.

Uh it it's already happening right there. I was just giving examples of the

there. I was just giving examples of the model solving open math problems and there are certainly robotic labs out there already. It's all kind of in the

there already. It's all kind of in the early adopter phase but uh at the pace that we're all moving I don't think it's long.

Um and it's so clearly the right thing for many fields that this is not you know we are not the only ones to have this idea. There are a lot of people

this idea. There are a lot of people that have this idea. There are a lot of interesting startups building things along this these lines and you know the world is just moving so fast it can't be long.

Right. Right. You shipped products used by hundreds of millions or billions of people. Um what was a product decision

people. Um what was a product decision you were nervous about that uh ended up being right.

You know the fun thing with uh product decisions at that scale is I any example I give I bet there are people in the audience who were like no no no that you got that one wrong.

very few are unambiguously right. Um

probably one was like ranking the Twitter feed which was extremely controversial back in the day. Uh

Twitter used to be completely real time and the the most you know the thing you saw at the top of your Twitter feed was the thing that was tweeted one second ago and the next one was the one that was tweeted 4 seconds ago. And if your

you know spouse or your best friend happened to tweet an hour ago like too bad you were never going to see it. It's

going to get totally buried. Um, but

there were a lot of people that said, "This is the magic of Twitter. How could

you possibly do that? You know, you're becoming Facebook now." So, that was very controversial at the time.

Although, it seemed in some sense like, "How could you not want, you know, you do care about different people's stuff more than other people's stuff? How

could you not want ranking if we could do it well and if we could bring the right balance of recency and everything else?" So, that was one. And I I think

else?" So, that was one. And I I think Facebook saw the same thing when they originally, you know, put out the news feed. You have a bunch of people that

feed. You have a bunch of people that are super upset, but then the metrics tell you an incredibly positive story, like double-digit positive kind of thing. And and so, you know, and then

thing. And and so, you know, and then you just you you you can just keep making that that better and keep getting wins there. Um uh it was interesting trying to figure out how

exactly we would um when we rolled out 01 preview the first reasoning model.

What was the right kind of UX paradigm for a um for a model that would not give you an immediate answer? Like all the other previous chat models, you type in an answer or you type in a question, you

basically get an answer right away.

There aren't a lot of experiences online where you have to wait like that. So and

and and the the model is doing this interesting thing with its chain of thought in the meantime u which we didn't want to expose completely because we didn't want to um because you can

distill that and you know basically copy our model which uh you know for for a bunch of uh geopolitical reasons we didn't want to have happen but you want

to show some so it was very interesting trying to figure out how you you was the model that people would go away and and just you know pop back in whenever it was done or were they going

to watch and if they were going to watch what would we show that and how do we balance not not giving too much information that would you know lead to model distillation but would be interesting so that was an interesting

um experiment and it's surprising sometimes building stuff with models a reasonable analogy for how should I you know handle the UX of this current situation is how would I want a human to

behave you know in a similar situation and Like if you ask me a question that I can immediately answer, you know, GPT4 style, then I will just immediately

answer, right? If you ask me a question

answer, right? If you ask me a question that I need to think about, I don't immediately start babbling my entire chain of thought, right? I don't just

like spew out whatever goes through my mind. I also don't like completely turn

mind. I also don't like completely turn around and go mute and like do nothing until I come back to you with an answer a minute and a half later. you know, I might I might say like, "Huh, okay,

that's an interesting question. Let me

think." And then you kind of give little like cliff notes as you think, well, it could be this. No. And so that ultimately is kind of what the model does, right? It it gives you kind of

does, right? It it gives you kind of like periodic updates of what it's thinking about as it's thinking. Um,

which is interesting in and of itself, and then it comes back with the answer.

So, we tried to model it a little bit off that.

One of the big debates like in terms of product development is data versus taste. What's a time when you said,

taste. What's a time when you said, "Hey, I've got a hunch. This is the taste. This is where we're going. We

taste. This is where we're going. We

need the a little bit more time for the data to catch up to it." If you just blindly follow the data, then it will take you like then you're not in control of where it takes you. And that's that I don't think that's where you want to be.

There's always a huge amount of value in anecdotes too when you get user feedback. Like the even though it it's

feedback. Like the even though it it's usually the case in my experience that if you're getting the data tells you one thing and you're getting a bunch of user feedback that takes you in a different

direction then what's actually happening is you have some biodal thing and what you're the data is like giving you the average of those answers and it's actually the case that you have two very different things going on

and you need to cut your data differently and dig in um because you should not dismiss the anecdotes.

the anecdotes are almost always valuable. I think the trick with data is

valuable. I think the trick with data is to understand it, not just be like, oh, you know, number go up like we should implement this thing. Um, but why does the number go up? Is it because we're

like because of novelty, which happens a lot, right? Some new thing that you

lot, right? Some new thing that you shipped, people are like, "Oh, that's interesting. What's that?" And so they

interesting. What's that?" And so they click on it once and your numbers look good, but they don't actually come back.

Or is it because they are confused? That

also happens a fair amount. either you

know you didn't get the product right or if you're you know growth hacking there are sort of negative states of confusion but they do make the numbers look good or is it like actually people are

retaining on this new thing and they like it and then you should lean into it. So you really want to like interpret

it. So you really want to like interpret the data trying to understand what it means underneath and then the decision is usually much more clear.

Yeah. So a lot of people in this room are building on top of AI. Um what's uh something that is there anyone in this room not building the uh a few a few but uh they're bu

still building with AI for sure but when you're building on top of AI what's uh a mistake you see uh startups and founders make one thing that we do internally that people sometimes that I see people not

doing is uh a lot of things today at least turn out best when you use um like an ensemble of models

If you have a hard question that you're trying to answer, you know, maybe it's customer service or something where there's a bunch of different things going on, people have different motives for um for writing in and you need to handle

their queries in different ways. There's

a bunch of actions that you need to take. The models are getting pretty good

take. The models are getting pretty good and now they really can sometimes just completely oneshot a tough flow like that. But you can you can make your odds

that. But you can you can make your odds even higher if you use models sort of together uh or you may have a an initial model that's orchestrating and is like putting a plan together and

understanding what you know what you should do to answer the question and then you have different models maybe some of them are cheaper models that are trained to do one thing really well um and the orchestration model is you know

calling the other models and things like I don't see people doing that enough I I think we're getting more sophisticated about it as as sort of an industry and at the same time as we're getting more sophisticated about it, the models are

getting better and so they need this kind of thing less and less. But that

still is an area where like behind the scenes we use ensembles of models in lots of places, trying to use like small models where you can, bigger models where you need to, and then have them all work together versus just like, oh,

let me prompt engineer one giant, you know, prompt and hope the answer is right.

Where's a a company or a product that's done a really great job of building on on top of OpenAI? Oh man, there are so many.

Something that surprised you maybe.

Well, I mean something that surprised me uh what like what's the new name?

Openclaw.

Yeah, openclaw built on codeex. Like that's

one of the most interesting things that has uh come out recently. More because

it's like a sign of what's to come. Both

like a the dude was able to put it together in the span of like three days, right? Which is it's just awesome. like

right? Which is it's just awesome. like

so many things are now possible in the span of days that would have been months and months of work or just completely impossible before. Um, but also because

impossible before. Um, but also because it points at like this interesting emergent world of uh of the AIs all working together. Have people spent any

working together. Have people spent any time on uh moltbook.com?

Oh yeah. So the it's uh you know you have all these AI uh openclaw agents that you know basically have access to somebody's full computer and they can do all sorts of things and you can command

them through you know messaging apps and stuff like that and now there is a social product for them called Moltbook where they go and interact with each other and talk about their humans and

tell stories and it's just fascinating.

I mean it's it's all weird. It's not

like most likely the next big startup or anything. It's just fascinating as a

anything. It's just fascinating as a sign of what's to come. Um I love stuff like this because like it just gives you a little bit of peak into the future.

Yeah, it's a the question is how much is emergent behavior versus how much is just novelty, right?

Yeah, it is a lot of novelty. There's a

lot of humans trolling through, you know, prompting their open claw uh agents. But there's also just really

agents. But there's also just really funny stuff.

If the product requires you to go out and buy a Mac Mini and you go out and buy it, then you've got product fit, right? So

right? So right at least among early adopter nerds like me we'll see where it like all right so you have a openclaw instance what are you sharing of your personal information

uh I am being careful you followed all the guidelines right I am being careful um there's but you know the tension there is like how do you have the full experience without sharing your whole life right so

yeah no easy answer no easy answer yet but I mean these are the things when when we first started building agents We were we had we were

very locked down because the last thing we wanted was there to be to you know even at low probability some instance where the the agent shared something it wasn't supposed to like it you know read

your GitHub repository and then shared private code or something like that.

It's like once you identify some of these problems that you want to solve models are getting very good postraining is getting quite good. It doesn't mean that we don't make mistakes but the the chances of a mistake are getting lower

and lower. you can build the sort of

and lower. you can build the sort of infrastructure and safeguards around it and I think people are much more comfortable now just you know giving more access maybe not all the way to

openclaw yet but um but just in terms of like connecting a whole bunch of different MCPs to various you know private data stores and then uh so like you you start with something like

openclaw where people are nervous but actually you know the the infrastructure kind of follows behind and patches up a bunch of the stuff you're worried about and then you really do get to do the the full promise also with good security and

you know again the pace we're all moving this this will not take long I mean it could be that open cloud's um uh legacy will be to accelerate the fully personalized you know agents

within like the skill players right yeah I mean one of the cool things about where we are right now is everything is new all the time like

today models can do something that computers have never been able to do in the history of computers Right? And like

in another month that's going to happen again and in another month it's going to happen again. And we don't know uh we

happen again. And we don't know uh we don't always know what's coming, right?

Sometimes these things are just emergent capabilities of models as we as we build them. Sometimes we do and we're like,

them. Sometimes we do and we're like, "Okay, we're specifically going to try and get better at this thing." But other times you're surprised, right? And when

you're surprised, all of a sudden you everyone in the world that uses the model has this new capability that nobody has ever had before. And then you have like the ingenuity of everybody in

this room and everybody outside these walls going, "Wow, what could you do with this capability?" And like we're all kind of discovering at the same time what you can build if you had these new

capabilities. So, one of the reasons I'm

capabilities. So, one of the reasons I'm so bullish about uh startups in general right now is like there's so many new capabilities and the world doesn't quite know how to, you know, what what's possible.

Um, OpenAI doesn't always know what's possible, right? we're not going to have

possible, right? we're not going to have all the ideas. Um, so it just is like the most fertile ground for startups that there has ever been.

Most of these tech shifts that we've seen like in like going back to the com era, it like the adoption starts with consumers and then goes into enterprise.

So this time around is different with enterprises first and there's not other than like open cloud there's not like uh a lot or or cloudbook there's not a lot of consumer oriented experiences that

are very native or you know video editing and photo generation. I was

going to say there's a little bit around some of that stuff, but there's not there's not tons. You're right.

There's not like where's the eBay, right? You know, where's the, you know,

right? You know, where's the, you know, the the the first generation, you know, big um consumer place. Why do you think that is? Um and do you think that uh you

that is? Um and do you think that uh you know, do you think that it'll change in the next few years? enterprise like B2B stands out because it's that is where we do the majority of our economically

valuable work and models are getting increasingly good at doing economically valuable work. So I just I think from a

valuable work. So I just I think from a where can you show value very quickly and also where is their money um uh B2B

makes a ton of sense because models also cost money to use. It's not like uh and and you know they're in a relative sense much more than like traditional you know get a database and pay network costs and

stuff like that. You you start having costs right away as a business in a way that maybe you didn't have as much if you were like building a consumer social thing before. Uh and so there's value in

thing before. Uh and so there's value in having early customers that can help defay some of those costs. And uh I don't know. I think it probably just

don't know. I think it probably just comes back to models being able to do uh economically valuable things in a way that without you know previously it was it was only humans that could do these

things. And so you can now build stuff

things. And so you can now build stuff in the enterprise and like take on you know save huge amounts of money or time or whatever for businesses. You can I mean we've seen how many how many

different B2B companies have grown gone like zero to 100 million to beyond in a heartbeat. So there's just like I think

heartbeat. So there's just like I think there's a bunch of lowhanging fruit there.

What advice would you give to a consumer startup founder that is thinking about distribution you know distributing through open AI now or in the future of viable path with the apps platform that we've built.

One of the ways that we thought about it was how or one maybe one of the success uh metrics for it is that you should see new startups being built on top of it that wouldn't have been possible to

build before. So it's not just if you're

build before. So it's not just if you're an existing company you can use it for distribution. It's actually it should

distribution. It's actually it should enable people to think completely differently about what a business looks like. Maybe you don't you can build a a

like. Maybe you don't you can build a a business in the future using this this apps platform that doesn't have a website or a mobile app and is sort of

entirely built around uh these kind of new platforms. Um so that's where we get that I mean that that's sort of you if you get there then you have an interesting platform. Mhm.

interesting platform. Mhm.

There's still a bunch of work. Uh it

we're still pretty early on that, but I'm excited about where it goes, especially as the models get really good at using a huge variety of tools and apps and other things. Like the the

value will be bringing all of that together in one interface and allowing you to do increasingly complex things by just, you know, typing into a a chat box

and letting the agents work for you.

Right. Well, thank you very much. This

was incredible. You've been a a great supporter to us by being here. your

company is the reason why many of us are here. Uh and also supporting the

here. Uh and also supporting the founders with a lot of great credits and and technical support. So everybody give it up for Kevin. Thank you.

Yeah. Thank you.