Cluely CEO Roy Lee on the necessity of virality | TechCrunch Disrupt 2025

All right, Roy, thank you for being

here.

>> Thank you. Oh my god, I hear that echo.

>> Yeah. Yeah. You get used to it. So, Roy,

you've gotten a lot of attention for not

just your product, but how you've

communicated it? Uh, you kind of have

this theory that distribution is

paramount, that it's important.

>> Yeah.

>> Can you kind of take me into your head

of when you figured that out and what

was going through your head?

>> Uh, yeah. I mean before Cluey there was

interview coder and I don't know if

everyone knows the story but it's like

the thing I built at school and um I

built it cuz like I I used to be the

biggest like YC Glazer. Literally this

time last year I was telling my

co-founder like bro imagine if we got

into YC and we we were like jerking

ourselves off over it but realistically

I mean it's like I tried a bunch of B2B

[ __ ] I'm like yo this is what Gary Tan

is looking for. But in all honesty bro

like like it was flopping. We didn't

make any sales. I couldn't even build

the thing and I'm like bro what am I

doing? Like I feel like I'm a I'm a

funny kid. there's other ideas out there

that are more viral and why don't we

just start thinking about like what's

the craziest LinkedIn post we could make

and uh that's what kicked off the idea

for interview coder like we're like dude

I'm pretty sure this can get a million

views and uh let's just post it and see

what happens and he posted that and

eventually it um it did get like a

million views and then we're like holy

[ __ ] this worked very well let's just

keep doing this and see where it goes

and eventually this becomes um like I'm

like I'm like front page of Twitter and

like like at one point bro like things

just start working so well you realize

like oh [ __ ] I'm looking at the frontier

of something. I got to see where this

takes me. And then um interview coder is

like the little translucent glass pane

that helps you cheat on le code

interviews. And we just thought like,

man, what if like I know we're not that

tough yet, but like what if we could

locally just make this work for

everything? Um and that's what sparked

the idea for Cluey cuz I'm pretty sure

like I'm pretty sure I just know how to

make a bunch of people see something. So

what's the biggest idea we could plant

in? And uh this is like a the biggest

idea.

>> Have you always been like this? Like

have you always liked getting people to

I mean that like in a nice way like like

cuz like you you've clearly cracked

something here.

>> So like have you always been wanting

like good at getting attention onto the

things you're doing?

>> My earliest memory um of all

consciousness is when I was 5 years old

in kindergarten

>> and we used to have like show and tell.

I couldn't read for [ __ ] back then. But

I used to bring like Korean books, like

novels, and I used to read them out loud

to the class and translate them in

English. But I was speaking gibberish

the whole time. And the whole time I was

just thinking like what is the craziest

story I could come up with to get the

whole class to like react about

something. So I I I say I' I haven't

really been in the space of content

creation for too long, but I've been in

the space of attention getting for my

whole life.

>> It's in your blood.

>> It's in your blood. Yeah.

>> Um, so I mean like you had interview

coder blew up. We wrote about it in

TechCrunch. I remember when it was

happening and uh kind of take me through

what happened like right after that.

Like you basically dropped out of

Columbia right?

>> Yeah. I mean, as it was going on, I

think everyone liked to say like, yo,

like you're getting kicked out of

school, like you really got to think

about it. Like, man, dude, you're you're

I'm so way too far deep for for your one

Twitter comment to be like, oh [ __ ]

you're right. Like, like now I got I got

to go back and and redo everything. And

like like you get stuck between a rock

and a hard place very quick. And as soon

as Well, Columbia sends me like

essentially, for those that don't know

the story, I I post about it. I post

about getting a job in Amazon um with it

and I make it public and it gets like

100,000 views and then Amazon and

Colombia are both like breathing down my

neck and as soon as Colombia sends me

the first letter saying hey you got to

go to a disciplinary hearing I knew like

I'm [ __ ] this product is [ __ ]

there's nowhere this goes but I have one

thing right now which I feel like nobody

else in the world has which is like I'm

the hottest thing on tech Twitter let me

investigate this space and see how far

it goes I don't even remember the

original question but um

>> that's good that was fine um So yeah, so

I I want to talk to you a little bit

because

startup marketing sounds really boring.

Like I kind of fell asleep a little bit

just saying that. But

>> you've made it like really exciting and

whatever you're doing, it doesn't feel

like marketing. Like it just feels like

these like viral moments that are just

like blowing up because they're

controversial. Like you're doing things

that other startups wouldn't. Like

people are outraged by it. People are

enraged by it.

I mean, what are people getting wrong

about startup marketing that you seem to

have cracked?

>> I mean, I'm pretty sure it's like what

people are getting wrong about social

well like like the original problem here

is like everyone who's good at

engineering, if you're any good at

engineering, you're probably not funny

and you're probably not going to be a

content creator because you don't have

it in your blood. And uh realistically

like uh uh most most of these people

have no chance of going viral. And thing

is like you don't need to like low key

if you build GBT then like you don't

need to go viral like like like it's

like it's crazy. But if you're building

like, you know, like an AI meeting

noteaker, then you probably want to

stand out somehow.

>> Um, and uh, realistically,

uh, what I think most people don't

understand is that social media has

extre.

And in order to stand out, you have to

be like very different. And what worked

in the past like I really don't think

Airbnb stunts of like uh the the Obama

O's or whatever like that wouldn't

really fly today because that's not like

every day people are used to seeing

crazier and crazier things. So to stand

out you have to be even crazier which is

why like you just see different content

on the timeline which is why everyone

gets mad and like upset like oh my god

like this is like some new new AI

disruptive like like evil like like man

everyone building like some random GP

rapper anyway. So it's like it's like

this it's not even that bad. Um, but but

you just have to frame yourself in a way

that is differentiated. I think I'm

particularly good at framing myself in a

way that's controversial. Honestly, I

don't even think I rage bait. Like

really, I think I'm just honest and like

authentic and I think people just see

that as rage bait because they they get

enraged, but uh that's probably just my

personality.

>> What do you think people find

controversial about?

>> Uh yeah, I I mean like like like from

the start like we say cheat on

everything. Yeah. And um this was a bet

like this was like

let let me actually think give you a

better answer. What do people find so

controversial about?

Um

I think I do a lot of things that are

different.

Uh and every single thing I do that's

different, I frame it through the filter

of my voice. And my voice just is

naturally very enraging to a lot of

people. So like like like like like like

like the way that I speak and describe

about things. I just have a hyperbolic

way of saying things. So if I say like I

could go out and say, "Yo, we're

building a desktop assistant that will

help you." Or I could say like, "Yo,

cheat on everything." And to me, like

that just sounds more like myself.

>> Yeah. And I mean I I think that

underlying, you know, your marketing

strategy, there's this notion that like

people are too worried about brand

safety, you know, that like if you go

and make something that like gets people

upset, that gets people angry about it.

I think a lot of brands are want to

steer clear of that kind of stuff

because then they have to put out an

apology.

>> It seems like you don't care about that.

>> Yeah. I mean, it's it's pretty obvious

to anybody who I think is under the age

of 25 and like scrolls regularly that uh

like reputation is sort of a thing of

the past. Like you can try and be like

the New York Times or whatever and guard

your ironclad reputation, but like

realistically you have Sam Alman who's

like on the timeline talking about

Hawkeye and you have Elon Musk like like

going batshit or crazy on the timeline

every And it's like realistic like how

could this how is this possibly the

world that we live in? And you just have

to realize that uh the world is trending

to a different place where you have to

be extreme, you have to be authentic,

and you have to be personal. And that is

uh extremely at odds with um being super

corporate. And I feel like there's a lot

of people doing this right now. Like

right outside there's a startup uh I

think artisan um that yeah like their

thing is like don't hire humans and then

there's friend who their thing is just

like you know here's a we're making an

AI device that you can be friends with.

So, like I'm curious like who do you

think is doing it well right now? Like

who do you think is taking what Kuli has

found and is doing it the right way?

>> The right way. I mean there's like Okay,

so I I I I guess there's there's two big

things that you have you need to worry

about and for all marketing your

marketing it falls in one of two

categories, very rarely both. And it's

either top of funnel or it's bottom of

funnel. Like it's meant to say, hey,

like you can either send the message of,

hey, look at us, we're clearly and get

the idea in people's heads. Or you can

send the message of hey download us to

get AI meeting notes and like like the

the the two are at odds. You can just

get more impressions and more reach by

saying crazy [ __ ] just attaching cluey

to it which is why you see like Stripe

powers like [ __ ] autonomous vehicle

data with with Stripe or whatever like

like like these billboards and it's like

I just just get get Stripe in your head

a lot and then later you see the one ad

that's like okay Stripe does this for

you. does like these easy easy payments

online and then you're like, "Oh, strike

the [ __ ] billboard that I've seen

like everywhere." Like, of course, I'll

buy it now cuz it's primed in my head.

And um I think if you only have one and

not the other, then it will be very

hard. Like like it's just not good

marketing. And um for you to have very

good bottom of funnel, like you actually

have to have like a product that works.

And um I think that's where uh a lot of

um the people in my generation

like fail is uh like like low key there

there there's only so many magic moments

out there and um yeah like like like at

at one point you can get all the top of

funnel and all the initial traffic you

you you you you want and at some point

people have to retain and people have to

like spread the product because they

enjoy the product.

>> Yeah. And I mean I want to ask you about

that because like the theory that you

have that distribution is everything

like I think clearly in a lot of ways

has become the benchmark for that

theory.

>> Yeah. Yeah.

>> So how's your business doing?

>> Um we're doing well. What I've learned

is there's like you should never I'm

pretty sure you should never share

revenue numbers cuz the only thing like

if you're doing well nobody's going to

talk about how well you're doing and if

you're doing poorly then everybody will

only talk about how poorly you're doing.

Twitter generally trends very very

critical and very negative. It's because

it's just a bunch of like jealous

engineers who like like don't want to

see anyone succeed except like

themselves. Um and and and and

realistically

I'll say like we're we're growing

quicker than I expected but it's not the

fastest growing company of all time.

>> Yeah. I think that's Merkore which is

they were on our stage like a day ago

and they they're sharing the revenue

numbers and you know it doesn't seem to

be going so bad for them. So

>> yeah, I'm I mean of if you literally are

the fastest growing company.

>> Well, there you go. Yeah. What about

like users? Like do you do you you guys

have like a monthly active user account?

>> We do, but I think this is just another

thing that if you share then none of

your other competitors will share and

all of them will just use that as a as

data against you. And um realistically

the second they do better like it just

gives them like uh like uh like fuel

fuel to the fire. And um I think there's

other ways to get impressions and gain

goodwill without doing that.

>> Okay.

>> Um

>> what about retention? Like because I

mean you're clearly getting a lot of

eyeballs,

>> but like do they stay?

>> The retention has gone up 3x from

launch.

>> Okay.

>> Um and that that's a number I'm

comfortable sharing, but I I also won't

share numbers here because I think again

like like it's just like everyone else

we benchmarked against each other. Um

but when we launched like the product

didn't work at all. Um, so like I I I

>> I guess it's good that retention went up

from there.

>> Yeah. Yeah. Yeah. Yeah. So, so, so it's

up 3x 3x from there and that's that's

what I'll share.

>> I mean, you've mentioned this before

that you basically created the launch

video for Cluly with just kind of the

idea and then kind of made the product

work after which actually I don't I

think you're just pretty honest about

that. I think a lot of startups do that.

But um, talk to me about like working

kind of backwards from getting

distribution and then

>> the product. Yeah, I think I think if

there's ah man I I I can't even say if

it's a mistake, but one thing that I

wonder is maybe we could have launched

too early because like real like just

for the first 3 weeks like even if you

downloaded it like you you you

realistically just could not use it. Um,

and right now a lot of the reason for

lower product velocity right now than I

would like is because we're suffering

from a lot of tech debt from a quick

launch and then now we have to maintain

backwards compatibility with all like

the random [ __ ] frameworks that we

added just cuz uh things were going down

so fast. And I think there is such a

thing as launching too early. Um, but

also like I mean we did like it's the

[ __ ] clue launch like how can I say

like oh I regret like being the most

viral like it did it did like it it it

it did did well for us. Um,

>> do you think you bit off more than you

could chew with at the launch video?

>> No, I don't think so. I think people

like people don't give a [ __ ] People

don't care about at the end of the day

it's like, oh, like it's like a good

good product for sending follow-up

emails like I'll use I'll use it for

that. Like at the end of the day, if you

just get like a couple thousand users

like they will find the use case for

you. And um that that that was the whole

idea is uh let's launch something that

barely works and then if we can get

enough initial users like they will

become the power users and they will

find out the use cases for us. And that

is what ended up happening. And now like

our our core power group of users is

just people who have tried it and

experienced like very different things

than we expected them to. It's it's the

whole reason why why we end up like

narrowing scope like well everyone's

using this for professional meetings

might as well just build it for that.

>> Yeah.

>> Um

>> so that's what the power users have told

you is that they just want to use it for

meetings.

>> Well they they they just do use it for

meetings.

>> Okay. Yeah.

>> All right. So that's what you're

building towards now.

>> Yeah. I mean it's it's it's it's the

it's the and this is another thing. I

don't know if this is inevitable. I'm

beginning to think that perhaps every

company that lands in a space, everyone

just sort of like gradually descends

down to the most optimal version of the

of the of the solution, which is why

like Instagram reels, Tik Tok, and

YouTube shorts all look the same. I'm

like per perhaps that is true. And

perhaps it is also the case that the day

clearly launched and um we were destined

to end up here.

>> Um whether we talked to a bunch of users

or whether we got super viral and then

and then and then use the usage data. Uh

I'm I'm I'm I'm not sure though. These

are still things that I'm that I'm like

figuring out.

>> So like now you're in the thick of it

like when the launch video went out like

you know and and you had some of like

the viral stunts where you like hired 50

interns to like just be influencers. Um

you know that was in a much earlier

stage. Now you're like okay you've got

some users you're trying to grow them.

Um, do you feel like the focus on

distribution, like the influencers that

you guys have, like the the focus on

social media is a distraction now from

actually building the thing?

>> No. No. And and and and the reason is

because you can just hire people to do

it for you. And um well, right right

now, I can spend a lot of my time like

not doing that thing and and and now the

marketing channels I can just like like

onetoone give to other people. Um, I

think this is something that every

startup probably needs to think about

more because there's just so many

startups that die because uh they they

could have PMF, but they they don't find

it because like the like like they can't

they can't get seen by users. In fact, I

know probably like four products off the

top of my head that like would be like

crazy like yo like lowkey's copy all of

their UI, but but they're not seen by

anybody and they get like [ __ ] like

less than 1% of the web traffic cuz

they're just going to get steamrolled.

And um I think like realistically most

startups do not die cuz like the the

product is not there. Like if you're in

a problem space and and you're you're

like solving something like even the

even the tiniest [ __ ] problem like

like there's money to be made there. You

die. You die because you can't find the

money. And um people just need a hard

focus. Like I think generally if you're

not in deep tech then you need a low-key

hard focus on on building distribution.

If your problem even remotely solve if

your thing remotely solves the problem

that you're trying to like you got to

get in front of people, bro. But like if

you guys have like millions of views

like on X, on Tik Tok, on Instagram, but

you don't have millions of users, like I

mean like talk to me about like

conversion cuz to me it's like if people

see it and then the product's good,

they'll convert into good into lasting

users theoretically, right? Like so so

why not just focus more on building the

product now cuz like you are still I

mean you might hire people to do it but

like you're in the Tik Toks like you're

you know clearly like still a part of

it, right? I'd say like the world is so

much bigger than people think and even

if you spend

uh okay at as an extreme if Mr. Beast

started any company and just gave you a

blank term sheet he didn't even tell you

the industry he just said yo I'm Mr.

Beast I'm building a company like what

post like he'd probably get like [ __ ]

10 like like a billion dollar post money

off rip just cuz it's like oh it's Mr.

beast. I trust him. So, obviously at an

extreme, everyone believes that if you

have enough distribution, then it

doesn't really matter. You can just

kingmake yourself.

>> Um,

>> and right now like like there's a big

like it might not seem like that in

tech, but there's a big jump like a huge

massive jump between us and Mr. Beast

and there's just like an enormous

enormous alpha to be gained there.

There's always like a best thing to do

and and certainly like I I spend a lot

of my time on product right now which is

why you see me tweet a lot less. Uh, but

like there's sure there's like a

globally optimal thing to do, but

there's like a hundred other really

[ __ ] great things you can do and

there's many roads to building a

generational company. And one of the

roads is could theoretically like maybe

the globally optimal strategy is just

[ __ ] spend the next 5 years Mr.

beasting yourself and then build a

company and maybe in that 5 years like

your your momentum in building the

distribution will just be better than uh

anything you could have done on product.

I'm curious like if everyone starts

doing this, if everyone starts doing

optimizing for viral rage bait content,

I know you said you don't do rage bait,

but like if everyone starts doing this,

do you feel like we end up in sort of

like a race to the bottom?

>> I mean, not really. At the end of the

day, like like you you can't fake

anything long term really. like at the

end of the day the the the the the best

products will win and the best marketing

will win and um if your marketing wasn't

the best then uh like you will die and

and I actually don't think that rage

like like pe if there is a globally

optimal way of getting people to convert

to the product then people will do that

and if it is rage bait then it will be

net better for the entire world because

it will mean that everyone will get

value faster um but it's pro it's very

likely that it's not rage bait because

things get saturated very quickly and if

everyone is posting rage bait then it's

not even bait anymore. Um maybe there'll

be like hope bait in the future and that

that will be what's viral.

>> So you have to level up now is what

you're saying like

>> you have to differentiate yourself

>> but this the kind of content you were

making before you're like okay now

everyone's doing that kind of stuff.

>> I really don't think everyone's doing

that stuff. My my biggest fear I I I I I

took I took a sort of like like a pause

from tech Twitter to to hard focus on

product because at one point we we had

everything we needed and my biggest fear

was like bro I gave everyone the [ __ ]

playbook like there's going to be

another royally and it's going to be

over for me and like bro it this didn't

happen bro like like like tech Twitter

the engineers just cannot make good

content bro you give them a [ __ ] like

a thousand tries on on on bad and they

they cannot make good good content. I

think right now it's not saturated

because like there's there's a bunch of

shallow replicas but I challenge you to

find one video who you think is like yo

this is as tough as cluey. I mean,

I'll take it. But, uh,

Friend,

like Friend had like a pretty like crazy

launch video that got a lot of people

like in New York right now, if you're on

the subways, like they have all that

rage bait there and it feels pretty

intentional, right? Like like is that

not the Cluey playbook or is that the

friend playbook? like I would check like

like what if like like imagine for a

second that friend actually worked and

it worked very well like

>> and and and it literally could replace

the function of like an actual friend

then do you like do you think people

would have the same reaction like

obviously [ __ ] not and as a counter

I'll say something that that that's been

like like like blow blowing up pretty

well is been like um

>> poke like poke by interaction theirs was

quite quite recent and um they use the

same launch launch like like like studio

that clue and they very arguably

followed the exact clue playbook, but

because the the the product was um uh

like like stronger and did did did

closer to the thing that they said that

they were going to do. Uh now all of a

sudden it's like oh my god like like

nobody even talks about oh my god you

guys spent like 250k on a marketing like

man like no nobody cares bro.

>> Yeah. I want to ask you about some other

uh companies and their like marketing

strategies cuz there was an injury one

of the investors in your from Andre and

Horowit said you're like one of like the

0.01% 01% of people who like really

understand distribution. So I want to

ask you about a couple. So like what do

you think about like OpenAI? Like what

what do you think of their like strategy

right now?

>> Yeah. I'd say like OpenAI is actually

running like another variant of the

Culie playbook which is when Sam Sam

Alman is on on timeline saying like yo

AGI imminent breath solved internal like

M. Okay. Okay. Let's be for real. But

like like he's he's doing this like like

drum up like fear and then all the all

the Instagram headlines go like yo like

Sam Alman just announced AGI and um like

like that's just another variant of

eyeballs on product selling a vision

that doesn't quite exist yet.

>> Okay. What about anthropic?

>> Anthropic um I think I mean if you just

look at like monthly active users

between OpenAI and anthropic it's clear

that one person is winning and I don't

think GPT5 is like gargantuanly better

than any of the cloud models. But

anthropic like also stirs up controversy

because they go on the headlines and

they're like, "Oh, AI is going to take,

you know, 50% of jobs in like five years

or something." So,

>> they don't do it nearly enough. And

>> I think they need to be doing that more.

>> I mean, yeah. I think I think almost all

content like is also the reason why

people will have like a super viral

launch video and then just flop like,

bro, like where's the next video?

>> And uh people just just won't have

enough quantity of of content. This is

also why like like there's been many

almost Andrew Tates since Andrew Tate,

but the reason why Andrew Tate won is is

like sure part partly because he's so

engaging with the way that he talks, but

also partly just because he makes more

[ __ ] content than everybody. Like I

don't know anybody else who's been on as

many podcasts and genuinely tried to get

clipped up that many times.

>> Okay. What about Donald Trump?

>> Donald Trump. I'm Bro, he's the [ __ ]

goat, bro. Look at him, bro. He's the

president of the United States, bro.

Like dude, if you think Donald Trump's

like like like in like the way he's

doing if if you don't think that's

working like how about you go be

president of the United States then?

>> I mean he so he's running the clearly

playbook though.

>> I I think everyone is running a variant

of the playbook of social media has

saturated most content formats that we

understand to be cool like 10 years ago

and as a result you have to

differentiate yourself and if the actual

core playbook is differentiate yourself

with content then yes Donald Trump is

running the same playbook.

>> There you go. What about the differences

between the platforms? Like Twitter is

different from Instagram and Tik Tok.

Like do you have to be like more

enraging on certain platforms than

others?

>> I I'll say there's like a few core

learnings and this might be like the

most valuable thing I say on here cuz I

actually think I have like a reason to

to speak on this. But um Twitter

generally is like super super caption

focused and it's not video focused at

like almost at all. The average scroll

time on Twitter is way faster than it's

going to be on Instagram and it matters

so much what your caption is. in tech

Twitter specifically, like people really

care that you're announcing even if

you're not [ __ ] announcing [ __ ] bro.

Just like act like you're announcing

something something revolutionary

because there's a lot of like high IQ

like like fake fake fake high IQ WJacks

on Twitter will be like, "Yo, like this

is interesting. I know all about this."

Um, so so focus on caption, focus on the

first frame that matters more than

anything. Um, on Instagram, Tik Tok, the

uh attention span, the time per scroll

is like is like slower. So you have

about 3 seconds to capture someone's

attention. Um, and in that like the

caption does not matter at all. The only

thing that matters is like how watchable

is the actual video.

>> Um, I'd say on Instagram things should

be vertical. Twitter things should be

horizontal. And also as a general rule

of thumb, the and this is something that

I like I I don't see a lot of people

talk about, but there's like different

levels of consciousness that you'll have

on different platforms. On Instagram,

Tik Tok, you have a very very low level

of consciousness because it doesn't take

that much to just sit down and watch a

video. Um, and as a result, that's why

things like mobile ad like like the

[ __ ] Flappy Bird ads like like like

those will do so well. is you're

operating at like probably the lowest

level of consciousness and it's just so

easy for you to be like emotionally

ticked by something and then click it on

Twitter because it's caption focused and

you have to actually read the thing.

You're operating at a much higher level.

Um it's it's it's not as high as if

you're like reading a book. But but but

but on Twitter this is the reason why if

you're building something like B2B or if

you have like like an actual value and

then like like like the onboarding takes

a lot longer and it's like a it's like a

desktop app or it's like a hardware

device like you're much better off just

solely focusing on Twitter. And um this

is true for any any any any any content

space that is like largely caption

first. It's it's true for Reddit. It's

true for LinkedIn. Um and I'd say

there's different businesses that have

no space being like many businesses have

absolutely no business being on

Instagram, Tik Tok just cuz the average

consciousness level of people there is

going to be too low for anyone to

actually convert.

>> So So you have to like be more in it. So

why don't those companies just like take

a page of the Kulie playbook and like

you know like be more enraging on them

like you think doesn't work for them.

>> I I mean I I I wouldn't even like again

I wouldn't say that enraging is the

playbook. It's like differentiated is

the playbook.

>> Okay.

>> Um for example, when when I said like

introducing the clue marketing team like

like that that that was a good video

that was just differentiated. Yeah. And

and it wasn't raging at all. Um

>> and I Yeah. Like I I I think if you can

make differentiated content then sure

like anthropics [ __ ] like coffee was

like like a different thing was why why

it went viral. And um people don't

necessarily want to see something

enraging, but the oh like obviously with

the platform like the reason you are so

incentivized to make something enraging

on Twitter is because on Twitter the

golden thing that you should care about

solely is quote post. If you look and

and this is something that is true like

I've reviewed a bunch of launch videos

from a bunch of founders and every

single time I see a bad one I will ask

them when someone quots this what do you

think they will say? And it could be

like you're launching GPT5 and it's

extremely obvious what they'll say like

everyone will jump jump in with their

take and um if you even like like like

you just have to imagine that it is very

very clear in your head when you quote

post it someone will say something about

it which is why you want to be like

enraging because then like like like the

angry opinions are the loudest but but

also like you can do this in many many

ways.

>> Yeah. Okay. So we've kind of gone

through a lot of the playbook here but I

want to hear what's next for Cluey. How

does Cle win?

Um, how does Culie win? I mean, right

now I I generally have this long-term

vision. I I I I think and like do do not

take me out of my word for this. I'm

just some random kid, but but I'm pretty

sure you have to hedge against both the

world of ASI coming relatively soon and

also the world that models just plateau

here. And if models just plateau here,

I'm of the opinion that among the

consumer AI tool space, there's not been

very many that are super interesting and

super value ad. And of the ones that

are, like I'd say like probably the AI

noteaker is like one of the only ones

that I've genuinely seen people gain

value out of. Um, so in the case that

the models plateau, uh, it feels to me

that that one of the most compelling

things is like the AI noteaker. So you

want to be in that space. We also have

to imagine like what if the models get

like [ __ ] ridiculously good. And if

that's the case, then you know, like

like the Cluey launch video was exactly

like like suggestive of if the if if

like GPT7 happened and this is what we

think that it would be able to do for

your life. Um and Cluey in its current

form factor, this translucent glass pane

that appears over everything and

augments the way you work on your

computer like like this is uh this is

primed to enter that space as well. Um

and I think uh the the way Culy wins I

mean at our core we're like an

application layer product. So uh it

depends on how well the models do. If

the models plateau here, then our plan

is to become the best AI note-taker

starting with the consumer. And um if

the models get ridiculously better, then

you saw the video yourself.

And I think that Clo, you know, is in

some ways become infamous for this

strategy. Like I think that a lot of

people look at it and say this is not

what tech startups should be about. You

should be focused on building the

product. You shouldn't be focusing on

just the marketing of it. I'm curious

what you say to those people, the

critics.

>> Yeah. I think for every single

successful person, there's like a

graveyard of other startups who have

failed and wouldn't share the opinion

that uh oh, you can just [ __ ] build a

thing and people will come and like like

like there's just so many people out and

usually the advice the only advice

you'll hear is from people who like Sam

Alman will say, you know, it's a lot

easier to raise $100 million than a

million dollars. Like yeah, if you're

Sam Alman maybe, but like like like

realistically most people if you're a

college student like like you you you

you won't think that way. And these

successful people um I think will will

say like what has worked for them is we

didn't really focus on marketing because

I was a nerdy engineer and I didn't

really know how to do it and I didn't

need it so you don't need it either. But

in reality like in their wake there is a

hundred other failed startups who if

they had adopted this sort of marketing

like they actually would have made it.

Um and you just don't hear about them

because they're not the ones that get

clipped up and they're not the ones that

are famous. Um, and I would say if you

actually looked at the net opinions of

every single person and at truly ignored

the noise of every single founder who

tried, had a chance and and failed, like

I I I I think a lot of them would say

like, "Yeah, low-key, I wish we had

better marketing." Um, and you wouldn't

hear the successful billionaire saying

like, "You'll just focus on product like

I did, and you'll be a billionaire."

Like, man, shut your dress up, right?

>> Who do you look to for advice?

>> For advice?

Uh,

it's hard. I feel like I'm I'm like uh

in like pioneering something different.

Um probably like [ __ ] like Jake Paul,

bro.

>> All right. Well, I think that's all of

our time. Roy, thank you so much for

being here. Thank you, everyone.