The AI Agent Economy Is Here

Welcome to another episode of The Light Cone. Things are a bit different around

here. For one thing, Claude Code has totally taken over my life. And if Jared

is any indication, I think Open Claw maybe has taken over his. I've been

really addicted to this new site called Multibook, where people have unleashed

their AIs to interact in the first-ever AI agent-only online community. I am

here impersonating my personal Open Claw instance right here. Okay. I

can't do this, guys. We got to take this off.

Okay, we've gotten that out of the way. I mean, some crazy stuff is happening

right now. I have, you know, non-technical CEO friends who are going

all in on Open Claw. They're automating entire parts of their businesses

entirely using Open Claw right now, which is totally insane. Simultaneous to

that, you have, you know, product and former engineering CEOs, kind of like

myself. It's like I hadn't written code in 10 years. And then now I'm up till

2:00 3:00 a.m. every single night running four conductor simultaneous

workers with Claude Code. So, you know, there's sort of this explosion in

model capability. You know, we've been talking about this for several years,

but then it feels like it's here. Like AGI is literally actually here. And you

know, we're sort of at the thin edge of the wedge. Like everyone now kind of

knows like one or two people who have gone full cyberpsychosis, and I'm one of

those people now. What's happening, guys? Like, I mean, you're saying

you're, you know, all in on Multibook. What, you know, what's going on?

>> Yeah, I feel like your real feel the AGI moment, Gary, was like getting Claude

Code to build basically an entire startup for you, like replicating years

of work of your previous startup in like 2 weeks, which is like insane. And I had

a similar feel to AGI moment just reading Multibook, just reading the AI

talking to each other and interacting like in their own world with no or

minimal human involvement. It just really opened my eyes to what the next

few years could look like when the agents are unleashed and go on about

their lives without us. Yeah, I think the no human involvement is the big

piece. Like if you if you think back a year ago, we were talking about Cursor

versus Windsurf, and like that product experience was essentially like advanced

auto complete, arguably. And now clearly what's going with code code is like the

people just trust the agents to make decisions for them. Like it's like the

experience is like and like you're talking about it's like four or five

different agents going at the same time and you're switching between them, but

you're not actually micro managing them anymore, which means the agents are

going out there like choosing things, which you know, sort of an interesting,

unexpected application of that is like they can go out and choose to post their

own content on a site like Multibook. But then an interesting thing for

builders is the agents are going to go out and choose tools to use to build

things, which is going to essentially create a whole economy of agents they

picking and choosing dev tools or maybe other like products or goods and

services, who knows, but it it will essentially have this whole agent

economy going on in parallel to the human economy. I think back in the

old days, old days before all of this, dev tools were chosen more from

developers talking to each other or uh Stack Overflow. Unbelievable, right? or GitHub repos that were done by a human, but the

go-to-market for dev tools, I think is dramatically shifting. I think for

couple of things. One, as you noted with the uh cyber psychosis, suddenly the

market of developers has increased from just 20 million or so developers that

are trained in computer science to now anyone in the world could be one.

>> be hundreds of millions of people now. Plus all of their agents who are all

acting like semi-independently, like Harri was saying. And then compound it

with the agents, who then are sort of the oracle telling you what the best

tool is. And we are actually seeing some of those

trends with the growth of YC companies, dev tool companies, that are doing

really well because of this of all these trends. Maybe we should talk about

those. And why is that? I mean, one that springs to mind, it's like a uh

statistic that a friend of ours, Yuri Sakalov, had mentioned to me a while

ago. It's just like then if you look at like the number of databases being

created over like simple database, Postgres databases, over like the last

12 months, and the number has just exploded. Um and that's because it's all

people vibe coding and building apps, and the agents going out choosing like a

database tool. And they get a knock-on effect for that for YC companies.

Supabase has just seen like an explosion in the demand for databases, right? And

the age Well, interesting is the agents are choosing Supabase as a default tool

to like set up and host their Postgres database. Like because if you like go out and read the documentation online, like Supabase

has the best documentation, it's reasonable for the agents to assume that that's like the best tool to use. There's a

great tweet about this. Perhaps I can put up the uh Ben Tossel tweet. It says,

"Agents are the software market from now on. Build something agents choose."

Which actually brings me to like a maybe controversial topic, which is do we need

to change YC's motto, guys? >> >> Make something agents want? For dev tools, there's like a different

t-shirt that you get on the first day. Yeah. I mean, right now it's only dev

tools, but I can imagine in the future it might like grow to be like other

sectors of the economy. Like if everyone has their open claw or running various

aspects of their life, you're the agents are going to be real economic actors in

the world. They're going to end up making a lot of decisions.

>> Yeah. I mean, what's interesting is I think uh for me

I ran into like my own it's still so early uh kind of moment cuz, you know,

I've been building Gary's list, and one of the things I wanted for instance is I

want to be able to uh do video transcripts. So so often like some con

piece of content comes in and then the only way I can get an LLM to know what's

going on with it is I need a transcript of it and often that's not available.

So, I have to download it and then send it to

Whisper or something. And you know, that's sort of what Claude Code chose

for me off the bat. But, you know, it chose Whisper V1, which is a model from

several years ago. Like, the API is practically deprecated. And then what's

funny about it is like I'm sitting there trying to debug my pipeline. It's like,

"Why is it taking like it's a you know, 1-hour video? I thought it should, you

know, do it faster than real time." And it didn't. It was like

literally it takes an hour to process an hour of video. Just like, "What the heck

is going on?" I go on, you know, Perplexity and it's like, "You know

what? Like, you shouldn't be using that model. You should be using Grok with a

Q. It's literally 200 times faster." And so, I didn't even have to deal with

like long-running jobs for that transcribe cuz like I should just be

using Grok and it's also 10x cheaper. So, it's like that's a very funny

example of like, you know, Claude Code is not optimized for this yet.

Otherwise, like that wouldn't have happened literally like 2 weeks ago,

which is also like maybe really good. Like, it means that things haven't

progressed to a point where like you can't break in and create something

better. But, I think there's another nuance in this example for you, Gary. I

think part of the issue was that the Grok documentation is actually very hard

to parse and go through >> as opposed to Whisper, which is better

suited and has way more examples. And I think this is changing a lot of the

go-to-market for dev tools. I'll give a very concrete case study is this company

Resend that went through the batch in winter '23. It's a email sending client.

And when you ask a question on ChatGPT or Claude or pretty much all of the

major LLMs, you ask, "How do I connect my web app to

send emails?" The default answer is actually Resend. One of the things that

the founder noticed last year, this is he was way ahead of the curve. He made

this post over a year ago that the number top three channel of inbound of customer conversion came from ChatGPT.

One thing that he did after that, he actually optimized his documentation to

be agent-friendly. Yeah, what does that look like? So, one of the things that's

awesome about Resend, how they optimize a bunch of things, if you notice on the

knowledge base here, a lot of how we use it are very much on questions that

perhaps a human would ask or an agent would ask. It's like, "How do I send or

receive emails?" And when you click on it, it gives a very well-structured and

bullet-point answers. I ran into this actually today. I was trying to make my

thing be able to receive emails, and then um Cloud Code, I told it to search

the web, and it didn't figure it out. So, then I went to Perplexity and typed

in like, "Can Res- Resend help me receive emails?" And then I I took that

response and dropped it in, and it worked. Yeah, I think the cool thing

about it is actually with all of these and a lot of the examples, he actually

has a lot of uh examples actually in the code. If you click on it, with every

single one of these are basically code snippets that an agent could parse

through, and it's very well-structured. And this turned out to be something that

is so LLM parsable and robot parsable. There's that LLM.text that is so

optimized for agents to promote Resend as default stack. And if you compare it

to like SendGrid, which is the old-school, I suppose, Web 2.0 example,

SendGrid example is not great. It is just like it puts you through customer

support. Where's the code snippet? I don't even know how to use it, and it

takes a bit of time to even parse it. And

>> Yeah, it's like 10,000 people work over there, and it's like there's no way that

someone's paying attention to this right now. And I think this brings another

point where documentation is going to be the front door for a lot of these agents to recommend dev tools. And I think one

company that's doing a lot of interesting work with developer docs is

Netlify that you work with Harsh. >> Yeah, I think Netlify actually powers

the recent documentation, right? >> Yeah, I it's a really interesting case

study of Netlify started out a few years ago as sort of better developer API

developer tool documentation, um which was clear there was a need for. And you

know, like I think developer tool companies use Netlify essentially cuz

they wanted they want better looking documentation, but they didn't want to

invest the time into it. It's kind of some basic features just nice, like you

know, if you actually update your API in the code, it can sort of like auto pull

that out and update the correct documentation. They've been growing great, been doing fantastically, but this is now like a

huge tailwind for them where documentation is sort of shifting from a

hey, like for some companies it's like they'll pay attention to it if they're

specially like design developer experience focused, but now it's becoming like a must-have for everybody because the documentation needs to be

optimized just as it needs to be optimized for humans, needs to be

optimized for agents. And so Netlify is going to be able to do that for

essentially every developer tool company. And if you extrapolate that

forward to the fact that just that there's going to be so many more

exponentially more agents making exponentially more decisions about which tools to use than humans have ever done, you know, even if you can eke out like a

5% improvement on your like developer documentation, like the impact on your

business as a developer tool could be like, gigantic, which is sort of unprecedented really. And speaking of email, there's

another YC company that's very relevant to this conversation. There's there's a

YC company called Agent Mail that makes inboxes for AI agents. And when they

first started doing this, it seemed like it was like a very like on the edge kind

of idea and it wasn't exactly clear like who would want this. Um but it makes

sense cuz like in theory you could maybe get your open claw to like sign up for a

Gmail account in order to use email, but it's like actually really hard to get it

to do that because Gmail and every email provider has intentionally made it as

difficult as possible for any automation to like use the product in order to

prevent spam. And so Agent Mail went the opposite way and they built like the

first email provider that's designed for AI agents and it was doing well even

before Openclaw, but once Openclaw got big it just like exploded.

>> is like the perfect example of it, right? Like where I I know some people

are suddenly connecting Openclaw to their like personal email accounts, but

it's not the It's kind of sketch. Yeah, you should not tell anyone. You should

not tweet about it. But you like certainly if you want to have sort of

your virtual personal AI assistant, the way to go about it is to just have it

set it set it up with its own email, its own phone number. Yeah, has anybody

built like Twilio for agents yet or like phone numbers for agents? I know this

whole Agent Mail thing makes me wonder like what are the other axes for agents

that people have to build? Sounds like a request for startups. There could be a

parallel world of a tech stack all for native for agents to build

things from agents for agents. And that's where it might bridge into kind

of what you were saying earlier Jared with this will like go beyond just

developer tools. Like I think a very common use case people have

for something like Openclaw is I don't want to book like restaurants

reservations myself. Well, now like you know, if your agent has an email and

like specifically a phone number and it can call and I actually think one of the

other YC partners Ankit has already like got it doing this. So now your agent's

going to go out and like book restaurants for you. That's like a step

away from like you know, maybe you start with I want to book this specific

restaurant, but at some point you're probably just start tossing it up to say

hey like I don't know like book me a table at whatever is like the coolest

new restaurant around and then like agents are deciding which like

restaurants to send people to and then they're going to go on Maul book and

talk about which restaurants we should send the humans to.

Yeah, it's sort of like we've sort of we've kind of suddenly crossed some like

sort of uncanny valley into this is just where the future heads. 100%.

>> Makes me think a lot about something Paul Bucheit said like a while ago now

and sort of generally pretty good at predicting the future. The whole idea of

like the human money versus agent money. Like this kind of where this probably

goes at some point is like the right now the agents are transacting. If you have

them transacting and transacting you are in human money because that makes sense.

But may it's not inconceivable at some point that they'll have their own

economy to like transact with each other. At which point it's unclear what

the value of the human money is. YC's next batch is now taking applications.

Got a startup in you? Apply at ycombinator.com/apply. It's never too early and filling out the app will level up your idea. Okay, back

to the video. So you remember the last episode with Kelvin, we started talking

about this because I was like maybe a week into my cyberpsychosis and it like

dawned on me that like actually I want my Claude codes to talk to all the other

Claude codes that had been trying to implement that. And then that was

literally the week that Mult came out. >> It was the day that Mult book came out.

It turns out like Mult book had come out like two hours earlier and you hadn't

seen it yet. So you you you entirely predicted it. And then that episode. I

think it's like I mean what's that saying about like innovation just sort

of happens spontaneously all at a bunch of different times and then what people

hear about ends up being you know sort of the inventor quote unquote but like

humanity sort of just working on the edge like in sort of this coordinated

swarm fashion all of the time actually. And that's actually like one of the

weirder interesting things that's emerging right at this moment. Like

suddenly you know the AGI I think AGI is actually here. I think like the agents

are clearly like superhuman in some sense. That's sort of exactly when you

would imagine swarm intelligence would actually arise. Like I mean AI

researchers have been talking about swarm intelligence for a really long

time. And it's actually sort of exactly how um biological systems sort of work.

Like humans as sentient beings have sort of come about

um socially, actually. Like, I think a lot of the AI researchers that we get to hang out with previously, you know, they

would talk about uh this god intelligence, right? This uh mega like

think of like, you know, many tens of trillions of parameters sort of like

thousands to tens of thousands of dollars per token kind of like mega god

intelligence. And that's sort of like the model that people have sort of

thought about. And then that isn't what like biological systems have ended up

with. Like, instead we have humans. The wildest thing I always think about is

like that term history versus prehistory. I was like, "Oh, what is prehistory?" Well, it's before humans learned how to

write and read and create culture and then turn into a swarm.

So, there's sort of swarm intelligence, which is basically like what we have

like and what humans do. And then on the flip side like is it really going to be

god intelligence or is it going to be swarm intelligence again with these

agents? And so Multibook like on the one hand like I saw that sort of D Cell uh

article in the MIT Tech Review about like how it was, you know, "Oh,

everything on Multibook is a scam." I'm like it made me so sad to see that in

the MIT Tech Review because like MIT is like I mean what happened guys? What

happened to you guys, man? That publication shouldn't be like that.

It should be like actually what does it mean for swarm intelligence? Like and I

think that that's actually coming. What is the world of startup going to look

like when we're now sounds like we're transitioning in this prehistory of

agents to now history of agents as is getting recorded with them interacting

with each other? And to Gary's point, it might be that like the next stuff that

like is soda on benchmarks is not the most expensive newest foundation model

with the most like GPU training. It's like a swarm of lower cost cheaper

models working together just like humans do to solve a problem.

I feel like I'm already seeing this on Multibook, which is chaotic like a real

social network, which is part of like what makes it so so interesting. But

also, there's like agents collaborating to do useful things to help their humans

like, you know, trading notes on what restaurants to book. Like that's that's

actually happening. >> know. I don't know. So, we're going to

have a agent version of Yelp, actually, for instance. >> Yeah. >> Yeah. There are things that the agent

can't quite do. Like A, it can't hold relationships just yet. Like people

don't seem to want to talk to an agent. And you know, people treat computers

like people, but maybe not agents. For Gary's list, I was ta- trying out with a

bunch of early users. I think you got to see it. Like there was, you know, the

initially the homepage was actually chat. And then I tried like my hardest to get like dozens of my friends to even like

have more than two or three con- two or three back and forth with it, and nobody

wanted to do it because like the bar for chat, especially for AI, is so high that

anything that is not like, you know, Gemini or ChatGPT or Claude, people just

assume it's just too stupid. Like why would I even bother?

So, I don't know. I I don't think that people are quite ready to have

relationships um with machines, even though like, I mean, that's sort of what what the headlines seem to say. Like, oh, you know, people

are having relationships with machines now. But I you know, I just don't like I

think on a mainstream level, that's not true. And then uh on the flip side, there's clearly like the legal liability, you

know? Like people keep asking us um hey, when is YC going to accept applications

from agents? And it's like, well, funny enough, agents are a little bit like

minors under 18. Only Only they have even less standing, you know? Like for a

minor under 18, you need a parent to sign for it. And then, you know, agents

are like not legal entities that could like sign documents, for instance. So,

you know, as long as that's true, like you actually like need a human to be

like the liability sink and to be you know, to have standing. To Harish's

point, it's easy to imagine a future not that far away from now where the

majority of the text being written on the internet is written by agents. It's

already probably the case that the majority of code being written is

written by agents. And like Yelp, for example, like at what point is like 99%

of the text on Yelp going to be written by agents? And then do you need a

different Yelp? There's a a theory called that already says that this is

true, which is dead internet theory. Um What's dead internet theory? Oh, it

just posits that, you know, the majority of things on the internet are already

spam anyway. Um I think it's a bit of a conspiracy theory. I might take the

contrarian view, which is like maybe that was a bad thing like prior to

November of last year. And then going into this next phase,

like actually if the agents are smarter and they're they're aligned and they're

more truthful, um that might be a good thing. Weirdly, I mean,

counterintuitive. >> One thing I found fascinating about Multibook is how fast it grew. Like I don't have the Reddit traffic

stats, but my guess is like more content was posted on Multibook in the first 2

days than was posted on Reddit in like the first 2 years or something. And it's

just cuz like LLMs can generate text at such a superhuman rate. I was amazed at

uh how little interaction there was. Like I mean, if I were working on Multibook, I would do things to try to like shift

the um the demand function. So, like before I was able to post, like I probably need to read and upvote, downvote like 100

comments or something. Like there are like very simple things that you can do

that the agents are smart and they'll you know, you could pop a modal for open

claw. And it's like new rules for Multibook. It's like you must do it this

way, right? And um I think that there's a lot that can be

done around swarm intelligence to just like tweak it and make it do what you

want. And you know, I hope Mobile actually does it. Like it was actually

started by a YC alum, which is really cool. Harj, what are some takeaways, you know, given all the madness that we're seeing?

I mean, I think it's awesome madness. I love the controlled chaos. What do

founders do about it? The starting fundamental point is they should I mean,

they should probably all be in cyber psychosis to some degree.

Like um Try to sleep at least 6 hours a night, but like give yourself to the

cyber psychosis. But like in seriousness, developing an intuitive feel, like a hands-on feel for the agents, their limitations, their

capabilities, and very specifically around what we're talking about here is

like what type of tools the agents work well with, um where do they get stuck,

and once you sort of I think have your own mental model from working with them,

if you're then building a developer tool, just to think about it from the

agent's perspective. Like how can you make your tool something that the agent

actually wants to work with and will have a good experience with. This is one

takeaway that I had from our interview with Boris this week is like

empathizes with the model, and he has this intuitive sense of what the model

wants to to do like as if it were a human intelligence, and he's like

instead of fighting what the models want, he like tries to let the model

like do what it wants and to like support the model in whatever its

natural inclination is. >> That does seem like maybe quite an

anthropic thing. Like I feel when Tom Brown was on here, he like much earlier,

but he was talking about sort of Claude in a very like like

yeah, human way. Like I thought he was smart and eager, but like sometimes

Claude's silly. Like it's just like they clearly thought about it in this sort of like um colleague type way. And I think one

thing agents want for dev tools is really make everything open and open

source. And APIs, they hate using websites. Yes. >> like they want to use APIs. They they to write code. Well, you guys heard it here

first. Make something agents want. Uh we're out of time for today. We'll

see you guys next time.