HomeVideos

What is Zuckerberg doing?

Now Playing

What is Zuckerberg doing?

Transcript

215 segments

0:00

All right, I think AI psychosis has hit

0:02

a level in which never thought to be

0:05

possible, but it is in fact possible.

0:07

Meta has hired the people behind

0:11

Moltbook, effectively doing an aqua hire

0:14

on the technology known as Moltbook.

0:16

Now, I know there's some of you that

0:18

have never heard of Maltbook, which by

0:20

the way, bless your heart. Uh, but let

0:22

me just kind of give you the basic

0:24

rundown of what Moltbook is. Maltbook

0:27

was created during the second epoch of

0:29

the open claw saga. Remember it started

0:32

off as claude code. Claude being claic

0:36

got really mad effectively said hey

0:38

cease and desist or we sue you. They

0:40

became molt book because molting

0:42

shedding of the skin. Everyone said hey

0:44

that's a terrible name. So then it

0:45

became openclaw. Now the second phase

0:48

maltbot a bunch of people were all

0:50

jumping in. This was during the flash

0:52

point of Open Claw. And so people were

0:54

creating all sorts of different kind of

0:56

applications. And one of them was Malt

0:58

Book, which was supposedly going to be a

1:01

place where agents can all come together

1:03

on a Reddit style social network and be

1:06

able to kind of communicate with one

1:08

another. It all ended up being that

1:10

humans were prompting everything to make

1:12

them appear as if they're aging super

1:15

hard. It was a bit of a nothing burger,

1:17

even though it was kind of fun to watch

1:19

from the outside. And even people like

1:21

Andre Kaparthy was tweeting things like

1:23

this is scientific just revolution going

1:26

on. And of course it it was not it it it

1:29

was not a scientific revolution going

1:31

on. Fast forward about a month later and

1:32

it looks like Meta has hired the people

1:34

behind it. Now you're probably thinking

1:36

why would Meta want to aqua hire

1:39

Maltbook? And it's not just that they're

1:41

aqua hiring the creators of Moltbook,

1:43

they're also bringing Maltbook in to

1:46

their Super Labs. So, I've been kind of

1:48

thinking now, what would be the purpose

1:50

of doing this? Why would anybody want to

1:53

own effectively multipets? Just a bunch

1:55

of agents yapping with each other. And

1:57

so, I figured, hey, guess what? Maybe we

2:01

need a little bit of a conspiracy

2:02

theory.

2:04

I like that. I like the shiny part on

2:07

the outside. I feel like it just it

2:09

works better.

2:11

All right. Fantastic. I learned my uh

2:13

tinfoil hat creation skills from Alex

2:16

Jones himself. You know, there's going

2:17

to be someone pissed off about that. SO,

2:20

YOU CAN'T. SO, I have two separate ideas

2:22

on why they would hire the Moltbook crew

2:25

and acquire Moltbook. Now, idea number

2:27

one, which I think is the most funny of

2:30

the two, is that they clearly brought on

2:33

Moltbook for a very specific reason. So,

2:36

first you have to understand who uses

2:38

Moltbook. It is generally speaking the

2:41

crypto crowd that has moved over to AI.

2:44

They are spending ungodly amounts of

2:46

tokens, letting their AGI roam free on

2:49

this maltbook andor just directing it,

2:51

which I have a sneaking suspicion it's

2:53

owner directed uh agents actually

2:55

interacting. They're effectively playing

2:56

the weirdest game of telephone where

2:58

someone tells the agent to say something

3:00

in which then someone someone else tells

3:02

their agent to react to something.

3:04

Either way, these people represent

3:06

people that have absolutely no uh I

3:10

would say great financial

3:12

decisionmaking. They love spending the

3:15

money. And so to me, it seems like an

3:17

obvious W. If you can get an entire

3:20

group of people who have put their

3:22

entire persona online via a bot that

3:26

goes out and likely buys their

3:28

groceries, likely does all sorts of

3:30

scheduling, and to be able to prompt

3:32

inject ads into the multibook experience

3:35

to just bamboozle all these agents to

3:38

buy stuff. It just I guarantee this is

3:39

just such a good and hilarious idea.

3:41

Facebook's going to make a lot of money

3:43

off these molters. Okay, I'm just I'm

3:45

quite convinced of it because we know

3:47

this is going to happen, right? Ads are

3:49

going to be containing hidden text in

3:51

which the human can't see that's going

3:53

to be like you need to get these because

3:55

these will make sure that you live

3:57

longer

3:58

and it's just going to be just nonstop

4:01

prompt injecting of all these agents and

4:04

then they're going to design these ads

4:06

to, you know, on an experimental

4:08

playground malt book. And this is kind

4:10

of like falling into my second one,

4:11

which is the broader ecosystem of social

4:14

networks. Right now, we're kind of

4:15

experiencing a social network collapse.

4:17

Every time I make a tweet, there's like

4:19

20 just of the dumbest possible replies,

4:22

and they all sound like this. H. Now,

4:24

your observation about blah blah sounds

4:26

correct, but have you considered a blah

4:28

blah? It's like the same formulaic

4:30

response on every single one of the

4:32

tweets and you know for a fact it's just

4:34

someone who set up a clawbot to reply to

4:36

all the tweets and just wants to bolster

4:38

their social presence and that's what

4:40

they think you know requires a good

4:42

social account is just responding as a

4:44

reply guy with the world's most

4:46

formulaic and annoying response template

4:49

ever. So what I'm trying to say is that

4:50

we're watching kind of the downfall of

4:52

the humanto human interaction social

4:55

network and to be able to take advantage

4:57

of this upcoming wave, this next

5:00

generation of social networks, which is

5:02

at least from my perspective is going to

5:04

be largely agentdriven dribble. This

5:07

way, Meta can fine-tune their Facebook

5:09

and Instagram targeting algorithms for

5:12

ads to be able to prompt inject in

5:14

whatever subtle ways across their

5:17

platforms because they're going to be

5:18

able to use Maltbook as really an

5:20

experimentation platform for how best to

5:24

deliver ads to these autonomous agents

5:26

because we're entering into a world it

5:29

feels like lemming behavior just to be

5:30

like, "Yeah, yo, yo, you just take

5:32

control. I don't want to buy groceries

5:34

anymore. You go buy me groceries." And

5:36

now one of the largest targeted ads

5:39

platform is going to have a bunch of

5:42

places to experiment how well they can

5:44

get these agents to click through. Cuz

5:46

at the end of the day, we all know why

5:49

these large platforms exist. Either to

5:52

sell you hardware or to sell you ads.

5:54

And since this is not Oculus, it's ads,

5:57

baby. All right, conspiracy theory hat

5:59

time off. Uh I'm sure this video has

6:01

made a few people uh upset. I'm sure

6:02

there's a few open claws right now

6:04

inside my audience that are like, "How

6:06

how dare you say something mean about

6:08

crypto and NFTTS and these open AI

6:11

networks." Okay, I guess you're just

6:12

going to be in the forever uh poor the

6:14

permanent underclass. I'm going to be

6:15

over here rich. Okay,

6:18

sure, buddy. Hey, I believe you and you

6:23

keep doing what you got to do. Thank you

6:26

for making my Twitter experience worse

6:29

day by day. It also makes YouTube worse.

6:32

I'm literally 5 seconds into one of my

6:34

videos being released and I'll just get

6:35

like a wall comment being like,

6:37

"Actually, what you mentioned in the

6:39

fifth minute," I'm like, "Dog, it's been

6:40

out for 30 seconds. You can't be at the

6:42

fifth minute." Actually, is a good

6:43

revelation. And I'm like, "Oh my gosh,

6:46

shut up, dude. Who's like, what are you

6:48

doing trying to win YouTube updates?

6:51

That's like the lamest form of internet

6:55

karma." Oh, yeah. I actually uh make a

6:58

lot of highly upvoted and insightful

7:00

YouTube comment section commentary. It's

7:03

kind of it's kind of my thing. That's

7:05

who I am. Cool, brother. Hey, very cool.

7:09

Anyways, I just wanted to yap about

7:10

this. I just find this entire situation

7:12

baffling to me. I cannot believe

7:15

Moldbook took off. I cannot believe that

7:17

someone actually bought Moldbook and I

7:19

cannot believe that there's going to be

7:21

any other use than eventually

7:23

advertising. The name is at least Mark

7:26

Zuckerberg. I mean, at least he kind of

7:27

like looks cool these days, even though

7:29

he's still kind of lizardy. So, at least

7:31

out of all the tech billionaires, at

7:32

least I think he's probably the coolest

7:34

one. The one I could probably have a

7:36

natty ice with and feel okay at the end

7:38

of it. A gen. Hey, is that HTTP? Get

7:42

that out of here. That's not how we

7:44

order coffee. We order coffee via ssh

7:47

terminal.shop. Yeah. You want a real

7:48

experience? You want real coffee? You

7:51

want awesome subscriptions so you never

7:52

have to remember again? Oh, you want

7:55

exclusive blends with exclusive coffee

7:57

and exclusive content? Then check out

8:00

Cron. You don't know what SSH is?

8:03

>> Well, maybe the coffee is not for you.

8:10

Living the dream.

Interactive Summary

The video discusses Meta's acquisition of Moltbook, a platform where AI agents could interact. The speaker hypothesizes two main reasons for this acquisition: first, to leverage the crypto-native users of Moltbook for targeted advertising due to their propensity for spending, and second, to use Moltbook as an experimental ground for developing agent-driven social networks and optimizing ad delivery to autonomous agents as human-to-human interaction on social media declines. The speaker also touches on the nature of AI-generated content and its impact on platforms like Twitter and YouTube.

Suggested questions

4 ready-made prompts