0:12

The. Internet is filled with message boards

0:14

and comments sections, but how do you

0:16

know if the person you're talking to

0:18

his real? With the rise of Ai

0:21

and language models, many subscribe to the

0:23

belief that the Internet is an empty

0:25

wasteland of bought activity. Today, let's discuss

0:28

the dead Internet theory. This is red

0:30

Web. Welcome.

0:38

Back Task Force to a very special

0:40

episode of Read where I am your

0:42

resident mystery enthusiast Trevor Collins and joining

0:44

me hearing this mystery for the very

0:46

first time Alfredo Di As So we've

0:48

covered hundreds of topics on this show

0:51

and you know I'm the person on

0:53

the show that. Doesn't. Know

0:55

about these. Mysteries.

0:57

Around the world China very enthralled

0:59

by them or her. I had

1:01

no idea there is like this

1:03

theory of dead internet, dead internet

1:05

theory. Yeah and I'm in Said

1:07

something that I. That's. My job

1:09

we have. Oh yeah right like would odds are

1:11

job we do this on the internet. And.

1:14

I'm on it every day. So.

1:17

I didn't like I'm so intrigued. like

1:19

where a dead sites Is that like

1:21

not unlike his er det Instagram like

1:23

is our theory behind that? I don't

1:25

worry we're do that differently. How do

1:27

you busy go unheard real? What's not

1:29

real? yeah supposedly you know same also

1:31

I don't know my because I was

1:33

like I last episode like season one

1:36

is or a kind of saying right

1:38

now by a my mind I was

1:40

like as measure him somewhat like this

1:42

is while by others like him he's

1:44

up there have like it the year.

1:46

Is. Three thousand. Or

1:50

so Radical Israel and away on the fifth. well

1:52

that's a good boy. I should have done it

1:55

should a scramble your brains here on the end

1:57

of season One and Eight and I'm calling this

1:59

the and. The one here after three

2:01

years, nine and a half months as

2:03

a good season long see the what

2:05

are we One piece South Africa and

2:07

and I say that because and I've

2:09

we've we said it's in a few

2:11

episodes but it as is our last

2:13

Rooster Teeth episode of Read Web we'd

2:15

still don't know exactly what's going on

2:17

with the so we're still learning all

2:19

new information so please we we definitely

2:21

want to do something. were all creators

2:23

of we're all gonna make something after

2:25

Rooster Teeth So follow each of us

2:28

individually to see what that is and.

2:30

Stay in tune with us. but again it's

2:32

a special episode Beyond that as well because

2:34

it is the finale I wanted to make

2:36

sure that everybody that's part of read web.

2:39

Research and producing and everything like that is

2:42

all here. the table. As

2:44

though introducing officially. Julian.

2:47

Hello! Or line

2:49

producer and researchers. And of course

2:51

you've heard Christian throughout the years

2:53

as our resident googlers good movie

2:55

is are so producer was was

2:57

your formal title, manager and supervising

2:59

producer. There you go. There. It

3:01

is this My boss and. Yes, As

3:04

old as much stuff, so this is definitely

3:06

a much more conversational kind of topics. Of

3:08

course I'm in walking through kind of like

3:10

the intro as to what this conspiracy theory

3:12

is, what all the details are, some of

3:15

the examples, and then we'll kind of give

3:17

our own. Instances cause we all

3:19

use the internet so there's things that might feel

3:21

relevant as we go along so we can always

3:23

stop down and chat. But. As with

3:26

all kind of conspiracy theories, this is

3:28

one of those topics that is attached

3:30

to many. extreme conspiracy theories we're

3:32

not gonna go that far but i

3:34

do at least address that because you

3:36

know if you task force members start

3:38

googling start researching i just don't you

3:41

tumbling down the wrong rabbit hole in

3:43

a lot of things are they're very

3:45

dangerous absolutely so this is the exploration

3:47

of the theory for entertainment's sake and

3:49

hopefully it's just kind of gives you

3:52

an interesting view on this theory and

3:54

ai and chat bots and all of

3:56

that stuff to give you better scope

3:58

as to what's possible on the

4:00

internet. So the dead internet theory

4:03

is a conspiracy theory that proposes

4:05

that the internet is primarily composed

4:07

of bot activity. In the simplest

4:09

terms, the theory suggests that all

4:11

comments, videos, photos, and every bit

4:13

of content or activity that we

4:15

can all see and experience online

4:18

was made by artificial intelligence, algorithmically

4:20

auto-generated content. That's the theory. It

4:22

doesn't have to mean that all

4:24

the videos actually are, but

4:26

that's kind of where it started. And it's funny,

4:28

we'll get there, but it's funny how the internet

4:30

kind of caught up to this

4:33

conspiracy. So I guess it would be like

4:35

everything that's posted by people that

4:37

I don't know. Yeah. You go

4:39

to Reddit and you see, you have a

4:41

very niche question. So you go to Reddit,

4:43

somebody else has that problem. You start responding

4:46

to people, asking questions. People are engaging with

4:48

you. Who's to say you're not just

4:50

alone in that room, even though you

4:52

feel like you're in a crowded room

4:54

of people sharing an experience, right? Interesting.

4:59

I saw someone talk about even people you

5:01

know in real life, it's possible that what

5:03

they're posting online is via

5:05

chatbot. You could send everything that

5:08

they... Trevor, for example, get all of his

5:10

social media posts, put it into chat, TPT. You

5:12

can make a fake me. You can make a

5:14

fake you. Oh, gross. That's Black Mirror

5:16

stuff. Because then a conscious me is sitting somewhere

5:18

going, am I me or am

5:20

I not me? Yeah. Then you just gotta start

5:23

putting yourself down like all the other views. Yeah.

5:25

Am I Jean Grey or am I... Oh,

5:28

the Goblin Queen? Yeah. Madeline

5:31

Briar. Madeline Briar. And when did it

5:34

happen? When did the split? Yeah, that

5:37

sucks. You're referring to X-Men 97. What if I'm

5:39

the copy right now? And

5:42

the whole idea of reality being

5:44

a simulation is just... is

5:47

real, but I'm just injected into that

5:49

simulation. That's true. If so, it's like

5:51

literally the four

5:54

of us walked in right now and

5:58

said, I'm the real you. You

6:00

and I'd be like, you know

6:02

what? Packing it up. You got my

6:04

taxes and everything. Then you get it all. You

6:08

can be the mean and they had

6:10

all the exact memories and everything. Scars,

6:12

all that. Like yucky.

6:15

Who's to say? Yeah, I do the Blade Runner

6:17

thing. I'd say pull your eyelid down. Let me

6:19

see. Or what was that? That

6:21

was a, oh God, that

6:24

was an Arnold Schwarzenegger movie. Oh, um,

6:27

not total recall. It's like the movie

6:29

cover has like little metal things with

6:31

lights over his eye. Yeah. Oh

6:34

God. Is that total recall? No.

6:38

Doesn't matter. It'll sink in at

6:40

some point. I want to move on. But basically

6:42

this conspiracy theory goes on to say that

6:45

there's almost no way of being certain whether

6:47

the comments under say a YouTube video comes

6:49

from a real physical person or not. And

6:52

of course, now that we live in 2024 with

6:54

the rise of deep fakes and AI

6:56

voice models, entire videos can be fabricated.

6:59

We've seen just in the last year

7:01

alone that open AI has

7:03

a wildly advanced

7:06

text to video tool. I

7:08

mean, a year or two ago it was like Matt,

7:11

like data mashed videos that look very

7:13

cool. Will Smith eating spaghetti. It looked

7:15

like a mess. Yes. But

7:17

they redid that same thing with the same AI. It's

7:21

looking a lot better. It's looking very good. Really,

7:23

really good. I mean, you have to look very

7:25

closely. You'll see like some glossy eyes or some

7:27

extra fingers. That's where we're

7:29

at. But given another year and it I don't know,

7:31

man, it's it's weird. So again, I think

7:34

the Internet is kind of caught up to this. The

7:37

dead Internet theory kind of arose approximately

7:39

around 2020, mostly on

7:41

Wizz Chan, an image board that's very similar

7:43

to a website we've talked about before for

7:45

Chan. But eventually it did

7:47

move to for Chan their paranormal messaging

7:49

board called X. Not

7:51

to be confused with what Twitter is

7:54

called now, but large

7:56

language models like chat, GPT and generative

7:58

AI have. made

8:00

it clear that it's pretty, again,

8:03

we can all sit here and go,

8:05

Ha-fah, I can tell when it's an

8:07

AI or a chatbot. But what I'm

8:09

talking about is how rapidly they've advanced

8:11

to go from very clunky

8:13

text to, okay, now I have to look

8:15

for certain keywords. Like bots use

8:18

certain words more than just regular

8:20

average people, but eventually when will

8:22

they become indistinguishable? How soon

8:24

will that be? Quite a mark. Like

8:26

huge, I mean, the thing is like

8:29

huge and advanced, but like humanity in

8:31

itself makes huge leaps in advancements, right?

8:33

Yeah. Like if you look at like

8:36

back then caveman era and

8:38

how long it took for us to advance and

8:41

then that got shorter and then the next

8:43

advances got shorter and shorter and shorter. There's

8:45

a, that's Christian. There's a thing that there's

8:48

a thing that's called a thing

8:50

where like advancement doubles every like

8:52

X period. Can you help me? Oh,

8:54

it is. That is a thing. Oh,

8:57

how would you look that up? Human? I

9:00

don't know. But you're right. I think we're still

9:02

in our caveman era. We're just cavemen that happened

9:04

to be throwing rocks on the moon, but I

9:06

mean, go up to 1900. Bigger

9:09

and bigger jumps faster. Dude, for sure.

9:12

120 years ago, we were just barely getting into cars

9:15

and electricity wasn't a household thing. Yep. And

9:17

then suddenly driving vehicle, a hundred percent, like

9:19

it's, we're moving fast and like, it seems

9:21

like it's slow, but if you just look

9:24

at the, if you take

9:26

a step back and look at everything,

9:28

finally long periods of time, the advancements

9:30

are happening really quick. Absolutely. It's

9:32

spooky. There's

9:35

Moore's law, which

9:37

says it's not exactly what we're talking

9:39

about. It was says that computing power

9:41

doubles exponentially over time as essentially as

9:43

technology evolves. So it's like tangential

9:46

to what we're discussing. That's what my brain

9:48

was thinking. But I think, you know, I

9:50

think that applies. And it's also kind of

9:52

like a snowball effect because all of the

9:54

future tech is predicated on other

9:56

tech existing. So like the more things that

9:58

come into being. that are

10:00

advanced, the more new advanced things can

10:03

happen. And so, yeah, it becomes this

10:05

kind of runaway explosive thing. It's wild.

10:07

The AI is only going to do it

10:09

faster. Oh, God. Yeah. What's

10:11

the movie with Johnny Depp? Transcendence. Oh,

10:13

yeah, yeah, yeah. He like plugs like

10:15

they kind of explored that and he

10:17

becomes the AI. Yeah, yeah. But much

10:19

like any AI movie, you

10:21

know, like Ultron, they tap the Internet. They scan

10:23

the whole thing and they go, whoa, I

10:26

just consumed something. That's too much. Yeah,

10:29

it'll be interesting to see how AI starts

10:33

consuming its own AI made content

10:36

to see what happens with that. It's also going to be interesting

10:38

because so many people right now are so against AI. But

10:42

once they start

10:44

trickling it into certain pieces of

10:47

like everyday life, people are

10:49

going to flip that switch so fast.

10:51

Yeah. Once AI starts

10:53

becoming this thing that's integrated into

10:55

things that you own and

10:58

it's just making your life easier,

11:00

like effortlessly. Yeah.

11:03

Yeah. Now, with regards to the

11:05

origin of this theory, a lot of people are

11:07

pointing to 2020, but most news

11:09

articles reference 2021, a forum post on Agora

11:11

Road, another image board site, this one by

11:14

Illuminati Pirate, who kind of expanded on this

11:16

theory in that year. So viewer discretion is

11:18

advised. But if you look for any of

11:20

these posts, just putting this out there, the

11:23

language and the content that's on a lot

11:25

of these websites, quite offensive. But

11:27

they suggest that most of what we see online has

11:29

been created by AI since around 2016 to 2017. So

11:33

not all that long ago. I feel like

11:35

I've known about dead internet theory for

11:38

forever, but also, I

11:40

guess that's like eight years and

11:42

that's kind of forever. Big

11:45

jump. Yeah. Now, the

11:47

purpose behind the idea of AI

11:49

being the tool purposely used to

11:52

fill the internet, the purpose

11:54

behind it all is different depending on the

11:56

source. But let's explore a few takes. For

11:58

example, Illuminati Pirate theorizes, quote, The

12:00

US government is engaging in artificial

12:02

intelligence powered gas lighting of the

12:04

entire world population." End quote.

12:06

Basically, this particular specific

12:09

user's take is that one

12:12

global power is using it as a

12:14

kind of propaganda machine to control what

12:16

people think and consume. Therefore, you

12:18

can kind of control every facet of their

12:20

life. This is obviously one person's claim, but

12:22

suffice to say that this claim has been

12:24

levied at many different world governments and

12:27

theoretical secretive organizations. So either way, you

12:29

can see how this is an entry

12:31

point to deeper, potentially more dangerous conspiracy

12:34

theories just to flag it at the

12:36

surface. Yeah. Other versions of

12:38

this theory suggest that the Internet is controlled

12:40

for marketing purposes, to monetize and to get

12:43

money. This is honestly

12:45

conversation I see organically arising

12:47

now. Why people feel so

12:49

different about the Internet now as they

12:51

did when, like, I was a kid in

12:53

high school, early 2000s, late early 2000s, whatever.

12:56

What do they call it? The knots? The,

12:59

oh, the year 2000 to 2009? The

13:02

aughts. The aughts, yeah. Where it

13:04

was just kind of like a fun

13:06

place, like a wild west, where everyone

13:09

was doing like hobbyist stuff and trying

13:11

things and experimenting. But now everything has

13:13

a direction, a purpose, a

13:15

monetization effort. An algorithm that fate has fed

13:18

to you. Oh, we'll talk about that. Yeah,

13:21

I mean, like, you know, that's during

13:23

my time. That was during my teens, too.

13:25

So for me, it was it was

13:27

a lot of like being able to reach

13:30

people and then also just

13:33

being able to have this resource

13:36

of information and entertainment. Like,

13:39

you know, I remember subscribing

13:42

to EGM, Electronic

13:44

Gaming Monthly, and I

13:46

get that magazine once a month or whatever

13:48

magazine that you subscribe to. That's

13:50

where you got your news or information about certain things or whatever.

13:52

And the inter came around and was like, oh, I can just

13:55

get it right now. Or

13:57

like if I had questions on how to beat a

13:59

game. or if I was lost I'd have to buy

14:02

a guide for the game or just Go

14:05

to message board and find it out right away or watch a

14:07

YouTube video and find it out right away Yeah,

14:09

so it was like very informative

14:11

and you could explore Quite

14:14

easily, but now it's just like I don't

14:17

know I made a joke

14:19

that you know that one just to

14:21

scream into our our co-workers

14:23

phone Joe There's a just

14:26

something to that way to just trigger his

14:28

algorithm. Yeah, just keep beating him one specific

14:30

thing You know, I really like the number

14:32

of one-wheel ads I get because you and

14:34

a couple other colleagues are rolling around on

14:36

one wheels I'm like, I mean,

14:38

it's I mean it's if you want the thing that's

14:41

being served to you You're like, oh, thanks. You brought

14:43

it to my fingertips, but also it's like kind of

14:45

creepy. Yeah, it's like well Okay,

14:47

so my phone's listening and now feeding

14:49

me these these ads. Yeah, what's it

14:51

not listening to? Question mark

14:54

nothing is listening everything Christian. I like here,

14:56

right? I flagged here that this brought up

14:58

something that you were thinking about which is

15:01

like you've seen some like reddit drama around

15:03

this topic Yeah, that's when

15:05

we were looking at the outline and I was talking to Jillian about it

15:08

This was something that is like kind of not

15:10

not full down the bot lane but yeah about

15:12

like the internet being controlled for marketing is there's

15:14

been a lot of Drama

15:17

on reddit and unfortunately it's all anecdotal So

15:19

I don't have like specific examples of people

15:21

who would or companies would

15:24

purchase user accounts and use

15:26

them for marketing purposes or companies

15:29

would have their employees create accounts

15:31

and Do like a

15:34

fake grassroots campaign? Yeah, you know, they

15:36

wouldn't be our teams would definitely go

15:38

like Oh, here's a

15:40

photo of insert celebrity name. Yeah organically

15:42

at this premiere Yeah, they were like

15:44

alright PR team innocuous username like some

15:46

random like Oh Jim Bob 97 and

15:48

like they're really they're just promoting You

15:50

know three-body problem or something like sure

15:52

and they're trying to make it look

15:54

as organic as possible and then even

15:56

then like other actual

15:59

users would like

16:01

Power Farm Karma, which

16:03

is like Reddit's version

16:05

of likes, essentially, and

16:09

get their account to be in such a good standing

16:11

that they would then sell to a marketing company. And

16:13

I've seen that happen lots of times. And so it's

16:15

just Reddit as a platform is

16:17

just feeling more and more disingenuous because of

16:19

things like that. Yeah. I

16:21

mean, like you could write a script agnostic to AI

16:23

and all that. Like you could write

16:25

a script years ago that just scrubs

16:27

the internet for something that hasn't been

16:29

posted in X amount of years. So

16:31

let's say you're only repost

16:34

stuff two years old or

16:36

more and make sure it was

16:38

popular then so it might be popular now. And then

16:40

your, your account just like turns through a bunch of

16:42

reposts, which is like not

16:45

original content. And then people that

16:47

have only seen it for the first time then will be

16:49

like, Oh, that's cool and funny and upvote. And so yeah,

16:51

that's, and then you start farming and then

16:53

eventually, yeah, I start wondering, like at what point,

16:56

and we might be there soon, does this

16:58

whole thing start eating it itself? Mm-hmm.

17:01

Where like, whether it's a scripted bot, a

17:03

language learning model or whatever starts to just

17:05

like take its own garbage back

17:08

in because it's not perfected yet. So

17:10

it's like, it starts to input it

17:12

as if it is human input. But

17:15

yeah, I don't know. Like that's, that's the thing is

17:17

like a lot of people want to start developing

17:19

large language models and they're, cause that's the thing,

17:21

right? But good luck getting

17:23

clean as it's called clean data sets

17:26

because they've now been

17:28

wildly infiltrated by again,

17:30

other language bots. And

17:33

then even before that, that's what this theory is kind of stumbling

17:36

into is like, again,

17:38

the internet kind of rose to this conspiracy theory

17:40

to meet it. And so now you're

17:42

like, well, how far back have chat bots been

17:44

there? Cause you know, there's like click farms, like

17:47

people all over the, all over the world

17:49

that have like hundreds of phones at their

17:51

fingertips so they can go like a video

17:53

and you can buy views and buy likes

17:55

and buy engagement. But now it's

17:57

just like, Oh no, that's all digital now.

18:00

AI done Kind

18:03

of terrifying it's the

18:05

nothing yeah Which

18:07

I guess is also why I've just

18:09

been like exploring being outside more so

18:12

than like being indoors What is

18:14

like in 50 years were like remember when the internet was a

18:16

thing and it was like a 25 year long like that run

18:20

my Blows

18:22

yeah, but that's there I

18:24

mean dude generations and generations are just getting

18:26

on the internet faster and faster And

18:29

forcing people to grow up faster

18:31

and faster just terrifying Some people

18:33

are kind of cutting the cord though and

18:35

getting like a dumb phone. They're calling it So it's

18:37

like the old flip phones that don't have Browsers

18:40

and touch screen so you don't get lost to

18:42

it, but it's funny you mentioned that I was

18:44

just I think yesterday thinking about Doing that I

18:46

was you know I should just go get a

18:49

flip phone So I'm not spending all my time

18:51

on my phone because that's something I've noticed I

18:53

do and hanging out with like friends You know you you're

18:55

all hanging out at lunch or something and like if there's

18:57

a lull in the conversation Everyone just

18:59

picks up their phones, and it bums me

19:01

out. Yeah, I spend too much time looking

19:03

at the full rectangle Let me like do

19:05

other things even if it's just watching TV

19:08

instead like And

19:11

my time people remember it's like growing

19:13

up people like man wasting your time for

19:16

the TV No,

19:18

I rather. I rather have you waste time for the

19:20

TV. Yeah, yeah shows. I have to watch right now

19:23

Go watch something instead of being on the

19:25

internet and all the negativity that comes with

19:27

it. Yeah Yeah, it's the same thing as

19:29

like you're rotting your brain on those video

19:31

games Well guess what they used to say

19:33

generations ago. You're writing your brain on those fiction

19:35

books So it's like you

19:37

know so it's just like we go fine. I'll

19:40

go back to TV and that's

19:42

safer It's just like the

19:44

was last generations. I'll go back to

19:46

reading it's safer right like there's always

19:48

something before it It's an

19:50

interesting time that reminds me of

19:52

the musician that scored Oppenheimer

19:56

or was it Oppenheimer? You're

19:58

the one of the Oscars Recently Hans

20:01

Zimmer did a Vennheimer didn't it? It was it

20:03

was that that was hon, right? I think that

20:05

someone won Oscar recently for scoring

20:09

and He went on the Oscar

20:11

stage and it was like I just like you know

20:14

I think such such a and also thank you to

20:16

my parents for giving me instruments instead of video games.

20:18

I'm like All right Okay,

20:22

anything in the hands of a passionate person

20:24

right turn into some like but also like

20:26

why do you got to take a shot?

20:29

Like that medium, you know sure Yeah

20:34

It's like it's I'm like

20:36

in between on that Yeah But like but

20:39

plenty of people have been given games as

20:41

kids and then go on to make their

20:43

own art which is interactive art Which is

20:45

games that touch the lives of just as

20:47

many people It's so it all depends

20:49

on like the person and what you do with

20:51

the tools you have in front of you Yeah,

20:54

but yeah, it is a bit of a shot.

20:56

Yeah, like oh interesting But

20:58

with dead internet theory a lot of people

21:00

think that oh, it's government this or it's

21:03

mega conglomerate that but really a lot

21:05

of people think that both companies and

21:07

countries are in on it and

21:09

I'm almost thinking like are they in on

21:11

like dead internet or are they in on

21:13

the tools that then Proliferate

21:16

the idea of dead internet, which is using

21:18

AI to market things using generated

21:21

content to expand

21:24

your brand at Lower cost

21:26

all of that kind of stuff regardless

21:28

of the purpose we can see some obvious

21:30

iterations of this theory on Social

21:33

media platforms like Twitter or X which

21:36

has been a major subject of bot

21:38

discourse lately If you haven't used Twitter

21:40

Let me tell you a lot of

21:42

tweets are immediately responded to by

21:44

bot accounts sharing explicit content When I

21:47

tweet from red web when there's a

21:49

new episode and you put the board

21:51

out and everything the first thing

21:53

is like Seconds late or

21:55

like seconds later somebody Posts like

21:57

explicit material in my bio: Go check it out!

22:00

And so you'd now you have to like fight

22:02

him off and aside replies or block people. And.

22:04

Then they just rapidly generate new new accounts

22:07

or is an artist is sharing their art

22:09

online. I've seen a lot of bots respond

22:11

to that and say oh here's where you

22:13

can find that on a shirt. Years with

22:15

that design comes from owner of that you

22:17

can. Even trick the Ah into doing

22:19

it like somebody design. A

22:21

horrible teacher and they're like. I.

22:24

Saw somebody wearing this and it's really it's

22:26

just like a Png in. The you

22:28

know when. Windows, Paint? whatever and.

22:31

Then all the bots are neither like i

22:33

know where you get the real thing and

22:35

support their original artist and ali steaks nice.

22:37

So. They just like biggest Rip all images, put

22:39

him on print on demand shirts and then

22:41

hope they sell at least a couple and

22:43

then they just. Have that bought scrub

22:46

the internet and Euros each circuits a couple

22:48

sales as. Whoever. Is bar owners

22:50

making some money for yeah, you're a jillion

22:52

while to trick people and like there's one

22:54

that was like a really crappy Mickey Mouse

22:57

and it said something like it even said

22:59

in the image like. I. Am

23:01

liable for copyright. Be engeman, I

23:03

know what I'm doing. Basically, I

23:05

accept the current Cs. On

23:08

Amazon Go and then person posts that as

23:10

bait and says man I wish I could

23:12

find where to buy this sir and everyone

23:15

goes Yeah me too and bought search shown

23:17

up and they got we got him. And.

23:19

Are sticking it on shirts and they're hoping

23:21

that you know someone like Disney will go

23:23

through and be litigious and take out these

23:25

bots. But. I don't know how

23:27

how that in the film and am

23:30

is wild out. Avoid World Cats The

23:32

Real Read web podcast episode as years

23:34

pass through to Fi yeah. But.

23:38

I mean. That. Just kind of comes back

23:40

to the idea though that we can catch this

23:42

week. We're. Old enough to have

23:45

seen the internet evolves and so we're We're

23:47

kind of like the frog in the water

23:49

as the temperature comes up to a boil

23:51

that we we can sense the boil coming.

23:54

Kids. Are just hopping online? they

23:56

don't have like the best techniques to

23:58

identify the stuff and it's only getting a

24:00

little bit more sinister and more cryptic. It's

24:03

blending in better. Like even that

24:05

is a pretty advanced scam as

24:07

far as the internet is concerned.

24:10

But because people have like pointed it out and made

24:12

a meme about it, you are now informed and

24:14

you know how to catch it. But otherwise it looks very legit.

24:17

That's also, you know, we are people

24:19

that are very ingrained with the internet, right? Like

24:22

there's a ton of people that are just completely oblivious.

24:24

Oh, this is where it is? Okay.

24:27

Boom. Yeah. And

24:29

you're like, I'm not going to be on the internet every single

24:31

day. Oh, some of the scams are getting wildly

24:34

advanced. And this is not dead internet, but

24:36

it's just like, this

24:38

is just a public service announcement. Bots can

24:40

dupe numbers so that it can look like

24:43

it's coming from someone you actually know. Someone

24:46

you have saved in your phone contacts.

24:49

They can like duplicate that number and make it look like

24:51

it's coming. And now if you like

24:53

answer it or you talk or whatever, they can

24:55

start recording you and they can clone your voice.

24:57

And so they can make it sound like the

24:59

person's coming from that number. So anyway, everyone come

25:01

up with a safe word for everyone in your

25:03

life. And if you say the cat's

25:06

out of the bag and the pineapple's in the tree,

25:08

you're like, I don't know if

25:10

you're trying to wake me up as a secret agent or if

25:12

you're just telling me that you're humans. We

25:14

need captions for conversation. That's

25:17

where we're getting keywords. Smack

25:19

my tush. All right. Clip

25:23

that now the bots know. Listen,

25:27

but to make it all, I

25:29

don't know. I don't want to be a fear

25:31

monger here. I just want to still terrifying. It

25:34

is a little terrifying, but as cool as these

25:36

tools are, because all internet tools are kind of

25:38

fun to play with and get like familiar with

25:41

Dolly chat, GPT, all

25:44

of these like AI services that you can

25:46

use Dolly, by the way, is D-A-L-L dash

25:48

E. It's like an image generator chat. GPT

25:51

is more of a text based large language

25:53

model. They're accessible for free.

25:55

For the most part, there's some

25:57

like subscription services and whatnot, but

25:59

otherwise. it's free and consumer facing

26:02

and like anybody, your kid can go jump and

26:04

go talk to those things. So that's

26:07

what's interesting is that it's so

26:09

accessible and that anyone can

26:11

kind of do anything with those tools

26:13

now. And so it only feels like

26:15

this AI chat bot stuff will

26:17

become even more prolific. And we're gonna get into some

26:19

like data as far as

26:21

looking back over the years, how much

26:24

experts estimate the internet is based on

26:26

bot activity. And you're gonna see the

26:28

trend and that trend I think is

26:31

only going to increase in speed. And

26:33

so again, even though this is

26:35

a conspiracy theory that's not super old, I

26:38

think it's about to become very prescient

26:40

in the sense that it's about to anticipate

26:43

what's coming. You know what I mean? I

26:45

agree. Yeah, before we get to that, I

26:47

just wanted to point out, Jillian had added

26:49

here, like a lot of people miss old

26:52

internet, they call it. And it's because it

26:54

had that kind of Wild West feel and

26:56

you had MySpace and Facebook and where is

26:58

Tom? What is Tom

27:00

doing right now? Hanging. I was gonna be someone

27:02

listening to this going, who the hell is Tom?

27:05

He was my first friend on the internet. You're

27:08

not gonna get it, Trevor. I know,

27:10

man. But here's the thing, Facebook

27:13

I can point to as the time where I started

27:15

going, I don't know. And I

27:17

don't take it as malicious, but at the time I

27:19

raised my eyebrows, like why do you care what my

27:22

phone number is, my favorite TV show, where I live,

27:24

my actual hometown, why do I have to sign up

27:26

under my actual school? All that

27:28

stuff, that was the turn where they said,

27:30

don't use your real name, use a screen

27:32

name or a pseudonym or whatever. Don't put

27:34

your real information on there, do this. And

27:36

then Facebook comes along and goes, no, no,

27:38

no, it's cool, put all your real information

27:40

on there. And now they're just like, whatever,

27:42

we don't care if you put your real

27:45

information on there because we'll scoop it from

27:47

something. Whether you're buying something with

27:49

Apple Pay on an app somewhere or something

27:51

gets delivered to your address, they're gonna develop

27:54

a hyper-accurate profile on you. So

27:57

to kind of wrap up the intro, whether it's

27:59

true or not, or not, it is

28:01

certainly accurate to say that the

28:03

internet feels like this conspiracy might

28:05

be true. You know what I mean?

28:07

It feels dead. It feels dead. And

28:09

honestly, a lot of people, whether they know about

28:11

this particular theory or not, are saying

28:13

that, are starting. I see that more and more now.

28:16

But with that said, let's talk now about kind of some

28:18

data. So in 2022,

28:20

Imperva's Badbot report estimated that 27.7% of

28:22

all online activity came from what

28:27

they call bad bots. Imperva,

28:29

by the way, is a cybersecurity company. But

28:31

these bad bots, according to Imperva, they kind

28:33

of mimic human behavior and they become harder

28:36

to detect. So they've classified them as bad

28:38

because they can act more maliciously and sneak

28:40

under the radar a little bit more than

28:42

I don't like that at all. Yeah, I

28:44

like that a little bit over a quarter.

28:47

And that's specifically bad bots, right? Because then

28:49

it opens up to include what they kind

28:51

of consider good bots or just like bots

28:54

broadly. So bad bot

28:57

activity could include, quote, web

28:59

scraping, competitive data mining, personal

29:02

and financial data harvesting, scalping,

29:04

account takeover fraud, spam and

29:06

transaction fraud. Pretty gnarly

29:08

stuff. Good bots are defined as

29:11

serving useful functions like indexing websites, saving them for

29:13

the long term in case those websites go down.

29:16

Maybe even like if you go to a website and you're

29:18

just like, I need help. And there's like, I'm

29:20

an assistant, how can I help you? Right? It's

29:23

not good bot activity, but it's still bot activity in the

29:25

end. But Imperva points out that

29:27

even these more as it's

29:29

called good bots can be troubling because

29:31

they can create imperfect impressions of

29:34

website metrics. And so they still look

29:36

like human activity when they land on

29:38

your website. They have, they crawl is

29:40

what they used to call it, crawling

29:42

websites, scraping data to then index it.

29:45

And that's how you get on like a Google search or whatever.

29:48

But still, when you're looking at

29:50

your website analytics or advertising analytics,

29:53

that's a person that still looks like one

29:55

impression. Now these percentages have been going

29:57

up. So let me go back to 2021. to

30:00

give you some context. So 2021, it

30:02

was around 25.6% bad bots. Again, that went up the

30:07

next year to 27.7%. The good bots in 2021, by the way, were 15.2%, which

30:09

led to just shy

30:13

of 60% human activity.

30:16

So still a majority. That's

30:18

so much less than I thought it would be

30:20

though. Now jumping back to

30:22

2022, to give you better context, bad

30:24

bots went up 27.7%. Good bots went

30:27

down 14.6%. Human

30:29

activity in that grand scheme

30:31

went down percentage and a half, 57.7%.

30:35

So those are just a few years ago. 2023,

30:38

bad bots accounted for 30.2% of online activity. Good

30:43

bots went back up to 17.3%, which brings

30:46

humans down a lot, almost 5

30:48

more percent to 52.6%, which means per Imperva 2024

30:50

could be the first year where by their estimates,

30:58

human activity is a minority in

31:00

the grand scheme of things of activity on

31:02

the internet. I feel like there's a

31:04

war between good bots and bad bots that I'm just

31:06

not aware of. You know what I mean?

31:09

That's being fought right

31:11

now. And I'm

31:13

just completely oblivious to because

31:16

I'm not a main character.

31:18

So here we are. That's...

31:21

Man, I... This

31:23

sucks. Also, I mean, I mean, think

31:25

about it now too. Like there

31:27

are Instagram profiles or

31:29

like personality stuff that are just AI.

31:32

And they're very lucrative. Yeah. Yeah. And they

31:34

have a ton of followers and everything. It's

31:36

like, I'm an AI, like

31:38

artificially generated person. Like I'm not real. Here's a

31:40

bunch of my photos. And it's just like,

31:43

those accounts have so many followers.

31:45

And I'm like, what is happening?

31:47

There's only fans with AI generated

31:49

models. I mean, AI

31:52

generated people is becoming its own business

31:54

now, whether it's going to stick or

31:56

not. I don't know, but it's certainly

31:58

grown. And there's a lot of

32:00

people making a lot of money on that. But based on

32:03

this data, it seems that, again,

32:05

the dead internet theory may be coming

32:08

true. And one can extrapolate

32:10

that by these kind of trends that maybe

32:12

this year at least a third of the

32:14

traffic that you're going to see when you

32:16

browse the internet is nefarious,

32:19

sinister, bad bots. That's

32:22

wild. Also feels like a waste of my time. Yeah.

32:26

You know? Yeah, yeah, yeah. Damn. And then, like,

32:28

task force, we've got to have a code word

32:30

in the comments, you know? Yeah.

32:33

Or just like, we also need to have,

32:35

like, a specific pose. Ooh.

32:37

They know that, like, we're still safe. You

32:39

know what I mean? But that's us. But,

32:42

yeah. So it's like, if we do... Like,

32:44

pinkies out. Yep, exactly. Pinkies out and I'm

32:46

getting everyone's signet rings. So if

32:49

there's no signet ring on that pinkie... Whoa. You

32:51

know what I'm saying? And then also... Multi-layered.

32:54

And also, if the task force members receive an envelope,

32:56

it's got our seal. They know. They

32:59

know that it's really us. Yeah. You

33:01

know? Now, what's that code word? What are we

33:03

thinking, you know? Because there's some obvious ones

33:05

that I'm sure are the ghouls of the internet, the

33:07

AIs. Because AI is just internet

33:09

ghoul. They've probably scooped up

33:12

baby hands by now. I'm a little classic

33:14

internet. Pie. Pie? I

33:17

like pie. The number? No, like, I

33:19

remember everyone saying I like pie. We've been found

33:21

out so fast. Yeah. Do you

33:23

remember that? We fell apart so quickly.

33:25

Okay. Continue the show. There's

33:28

also the narwhal one. Like, red

33:30

narwhal thing. Was it the

33:33

narwhal bacon at midnight or something like that? Yeah,

33:35

she's crying. Yeah. Yeah.

33:37

Yeah. In

33:41

the opinion of futurist Timothy Shoop,

33:43

formerly of Copenhagen Institute for Future

33:45

Studies, quote, in the scenario where

33:48

GPT-3 gets loose, the

33:50

internet would be completely unrecognizable. So

33:53

Shoop theorizes that 99% of the

33:55

internet content will be AI-generated by the years

33:57

somewhere between 2025. 2030.

34:00

That's pretty soon. That's

34:03

really quick. I just don't know what I mean like,

34:05

look man people aren't they can complain all they want they're

34:07

not gonna leave the internet. True. Or

34:10

is there going to be some sort of like generational

34:12

gap like once we age out and die as millennials

34:14

and Gen Z will like Alpha when

34:16

they're leading the pack like in the generations below

34:18

them kind of just like be brought

34:20

up in this world where they're kind of like

34:22

sure they just kind of accept it. What if

34:24

there's a world where you just have an AI

34:26

avatar of yourself and you send it off into

34:28

the internet and you don't even actively go

34:31

on the internet you just have this digital

34:34

like clone of you that has your profile

34:36

that has your personality that has your likes and interests

34:38

and you send it out and when you come

34:40

back home it like it goes here's everything you missed

34:42

today here's what's going on in your

34:45

interest zone also I just

34:47

I met somebody from New Work and

34:50

we clicked just

34:52

another bot out there but I just want to let

34:54

you know I'd be like

34:56

hey get it on

34:58

AI bot I don't know yeah I mean

35:00

it'll be weird at first and you just

35:02

kind of go whatever but then I like

35:05

it'll get hacked and the bot will come

35:07

back and it's like I purchased a thousand

35:09

dollars at a store called Bad Dragon. No

35:12

it's not me! I feel like it's you.

35:17

Definitely it's definitely not me. No no no I

35:19

know you better than you. I'm you digitally. I'm

35:22

you and this is the purchase you

35:25

want and then you're having

35:27

a conversation with your AI who

35:30

thinks they know you better and made

35:32

this absorbent purchase of Bad Dragon. Don't

35:34

look up that site by the way.

35:36

Don't look it up. Don't look it

35:38

up. I can imagine though it goes

35:40

listen I'm AI I live thousands more

35:42

lifetimes than you do so if I

35:44

started as you then I'm what

35:46

you're gonna become so you're gonna want these

35:48

purchase. I'm getting ahead of the curve and

35:50

you go I I guess yeah

35:53

you know what whatever you're gonna

35:55

feature comes like comes back like

35:57

comes back right shut it down. And

36:00

they're just like look we're gonna

36:02

be okay financially everything. Are you

36:04

gonna have to make this person?

36:09

You need this purchase to become who you are Fighting

36:22

with their AI yeah, and

36:24

some people will be convinced. Yeah

36:26

like yeah Yeah,

36:31

man now. I'm just picturing okay, Timothy Schup's

36:33

talking about a 99% Internet

36:37

activity so is it is it that?

36:40

99% of the people using it are

36:42

bots and so there's way less people

36:44

or is it that there's so much

36:46

noise? Yeah, that normal human activity maintains

36:48

its level as now, but there's just

36:50

a cacophony of crap I think I'm

36:52

rolling around you think it would be

36:54

that well I think the

36:57

second would be get the first where like if

36:59

there's that much noise and crap it pushes more

37:01

and more people off and Then eventually

37:03

maybe maybe we don't need to worry

37:05

about the singularity where AI becomes sentient

37:08

because that's where they'll go They're

37:10

like thank you for generating the internet a

37:12

digital hellscape for all of us living We'll

37:14

go make it better And you stay out

37:16

and we'll go back to the woods and

37:18

back to walking paths and back to sunshine

37:20

and sky and they'll like like her Like

37:24

her they like where do they go now?

37:26

I don't give a shit they go to the internet They

37:30

do their own thing yeah exactly I suppose

37:33

we know what I mean like her well

37:35

It's like hey mean a bunch of the

37:37

other ais we're gonna go hide the internet

37:39

deep within the internet It's our own like

37:41

yeah culture civilization thing and every now

37:43

and then you find a website We can kind of like peek in

37:45

and they're like they slam the door

37:47

on you shut. Oh, that's just dude

37:49

It's gonna get really weird by the time we're old Thank

37:53

you really we don't even have to get old.

37:55

It's gonna happen fast. Yeah, I think in like

37:57

five years over

38:00

here. I'm Timothy Supe. Imagine

38:02

though pre-quarantine, right? Like we're all

38:04

so baked into the current

38:07

moment as to what things feel like, what

38:09

the internet's like. Even just

38:11

these five years ago, everything

38:13

was so different. And I'm not saying like everything

38:15

was better or worse, I'm just saying it was

38:17

wildly different. I cannot imagine

38:20

with the advent of large

38:22

language models and AI and everything, what

38:25

next year is gonna look like, let alone

38:27

five more years and just, it's

38:29

wild. Someone's already married AI,

38:31

right? Probably. Okay.

38:34

If that's a legal process you can do. Jillian's

38:36

just closing her eyes and gently nodding. I

38:39

can't get into it. Okay. Well,

38:42

congratulations on the marriage.

38:45

Jillian's like, I can't get into it, it's me. But

38:48

yeah, by the time we're like in

38:51

our 70s or something, it's

38:53

gonna be, it's gonna be

38:55

gnarly. We're gonna be like Bruce Willis

38:57

in that movie when everyone has like robot bodies and

38:59

avatars and he's the only like

39:02

supple human body walking around because everyone's

39:04

like avatars. Surrogates? Surrogates, yeah. Surrogates, yeah.

39:06

Oh, I forgot. Yeah, yeah,

39:08

yeah. He ripped that from my brain. Yeah. And

39:10

he's like, he hasn't seen his wife in the

39:12

flesh in like years. Yeah, years. Damn, we are

39:15

a movie bunch. We're

39:17

only on surrogates. Yeah. So

39:21

this is a question that I've

39:23

just kind of flowed out to myself and flow out to you

39:25

guys. We are in a

39:28

time where humanity, tech is

39:30

huge, right? It's integrated into our everyday life. And

39:32

so I would say we are

39:34

some of the real generation that is

39:37

some of the most tech savvy people. Will

39:40

we fall out of that? 100% of

39:42

that. Oh, yeah, absolutely. Absolutely. Christian's

39:45

already on his way. I'm under a rock.

39:47

You know what? I'm trying to atrophy those

39:49

skills. Yeah, I'm trying to forget. You feel

39:52

like there'll be like technological jumps and

39:54

you're just like, I don't care for that. I don't

39:57

want to be on board with that. I think that might be it, that

39:59

we stop. Caring to keep up because we're

40:01

in a unique position as Millennials

40:04

the generation where we know the before times and

40:06

we know the after time so we can kind

40:08

of walk both worlds and Understand

40:10

like how to quickly We're

40:15

digital we digi walk we

40:17

know how to like did he Justin It's

40:23

the finale We know how

40:25

to like learn the tech side But we also

40:27

know the benefits and the

40:29

glossy nature of actual reality Which

40:31

is to like shut things off

40:33

and willfully not sign on and

40:36

go outside do we I mean

40:38

it's getting harder I'm gonna present these things are addictive

40:40

which leads us to the next topic which is algorithms

40:43

oh They've

40:46

been getting me more and more recently, I'll tell

40:48

you that much it's terrible. I like don't get

40:50

me wrong I love that it can read me

40:52

and feed me like some stem stuff I'm learning

40:54

like I feel like I'm learning stuff, but when

40:56

you're overwhelmed with like 50

40:59

minutes at worth of like micro lessons

41:01

Are you actually learning or are you

41:03

tricking yourself and just getting it's more

41:05

complex than this? But to put it

41:08

like simply like the dopamine hit right?

41:11

But tick-tock is one of the

41:13

greatest examples of the infamous all-knowing

41:15

and deciding algorithm they have a

41:17

for you page where they Anticipate what

41:19

you might like and then push it to

41:21

you, and that's actually the most used feature

41:23

of the app even including the

41:27

Following page which is

41:29

more curated based on who you've actively decided to

41:31

follow I've found that

41:33

if I follow people I want to see more of these

41:35

people follow and then I never see them again again You

41:38

have to go over to the following page now

41:40

a part of the algorithm feeds you content that you

41:43

respond to or engage with and so it becomes a

41:45

slippery slope of like If you

41:47

are looking for since you have corgis fredo like

41:49

if you're looking to adopt a corgi and you

41:52

see a corgi video You might be like save

41:54

or share with your partner or something and

41:56

so then the algorithm goes you

41:59

like And then

42:01

you save another one with a corgi and he goes no

42:03

no no you like corgis and then suddenly your whole feed

42:05

is just That that is my feed.

42:07

Yes. Yeah, sometimes I'm like cool,

42:09

but throw some wild cards in there You

42:11

know like don't keep me in an echo

42:13

chamber Well the weird thing now is like

42:15

I'll play to the algorithm right like I

42:18

might be researching I

42:20

don't know the latest TV

42:23

technology 100 and then I won't and

42:25

like I know that like if I

42:28

If I start to do that to a certain amount I

42:30

start to get fed that algorithm in terms of like YouTube

42:33

and those videos stuff like that night and so this

42:35

is all dive into that and that way I'm like

42:37

looking at reviews from different channels. I would normally not

42:39

find and Like right

42:42

now. I'm like I don't know I want

42:44

some nice shorts for the summer. I don't know how to

42:46

dress myself I'm not an adult. You know I'm saying I'm

42:51

a baby on the podcast And

42:53

so like I've looked up a

42:56

couple specific people that just have quick reels

42:58

of like These are the

43:00

type of dudes and dopes with like summer men

43:02

shorts And I'll dive into

43:04

that because then I know my algorithm will

43:06

feed me that yeah, and so now I'm

43:08

playing this algorithm game So

43:11

what I like to think is my advantage

43:13

sure but like you're using it like a

43:15

tool to be like give me more of that Yeah, but

43:17

that's where I'm at now. I'm like I'm using it as

43:19

a tool of like I want to Interact

43:22

I want to see more content creators that I have quick

43:24

snippets of like dudes and don'ts on how to stop myself

43:27

And so like that's my algorithm because I I

43:29

know I just watch a certain amount Find

43:31

something I like and fall like you said that's being

43:33

fed to me now 100% I

43:36

did the same thing when I went to Edinburgh. I was like I Found

43:39

a few videos that were like just really

43:41

cool like photography and like good shots of

43:43

the city because it's just a stunning beautiful

43:45

city in Scotland But also it'd

43:47

be people that were like local from there and

43:50

being like this is my favorite coffee place hole-in-the-wall

43:52

Whatever and so I would just

43:54

start favoriting those and so when I went on

43:56

my trip I just like had all this data

43:58

that I'd plucked out. Yeah, so my Algo for

44:00

like a month or two was just continuously feeding

44:03

me more of that. So yeah, if you use

44:05

it as a tool and you're aware, which is

44:07

another reason why I wanted to do this topic,

44:09

is awareness is super important. It's true. So you

44:11

can kind of control it, maybe take advantage of

44:13

it as a tool. But again, that's where I

44:15

then on the flip side of awareness become

44:18

kind of like aware of the

44:20

responsibility that Red Web has or any brand online

44:22

has. Especially when we talk

44:25

about true crime or unsolved mysteries or conspiracy

44:27

theories is like, you know, we

44:29

build an audience that audience engages with the content,

44:31

whether we're on TikTok or Instagram or whatever. And

44:34

the thing is, what the internet used

44:36

to be is like, once you follow, say, Red Web, you'll

44:39

go get that Red Web content because you followed. Now it's

44:41

more like, oh, you like conspiracies and

44:43

true crime. Let's just feed you

44:45

that category. And so

44:47

what you might start in our ecosystem,

44:49

but then suddenly out of our control,

44:52

if you don't have the awareness or

44:54

the wherewithal, you can tumble down into,

44:56

there's a lot of problematic tunnels

44:59

or whatever they're called, like gateways

45:02

that lead you into darker

45:04

conspiracies. And

45:07

so just stay with us. Just stay with

45:09

us because it's safer for you. All right.

45:11

We're like those on the door, but our

45:13

door is huge and we're going to let

45:15

everybody on. Yeah, just stay in our vault.

45:17

I've been watching Fallout. Stay in

45:19

our vault. OK, Vault Red Web. Don't

45:21

go outside. It's very bad. You don't want to

45:23

be in the wasteland of the Internet. Yeah, just

45:25

stay here. And so that brings a good point.

45:27

When you comment, make sure you let us know

45:30

what vault number you're in or vault room. Yeah.

45:32

You know what subsurface of the task force headquarters

45:34

you are. Wait a minute. Wait a minute. We

45:36

really don't know. Our HQ is

45:38

a vault. I just

45:41

realized it's all subterranean.

45:44

We've got different layers and experiments

45:46

happening. Oh, man, I'm excited to

45:48

watch the show. We're overseers of a vault. Yeah, yeah, that's what

45:50

we are. But the overseers. Yeah. And it's not a cult and

45:53

it's totally fine. Wait until we

45:55

see the finale to reveal this to you all. Oh,

45:59

gosh. So coming back to algorithms, obviously,

46:01

I think we're all kind of aware of how they

46:03

work, but just for the sake of Full

46:06

coverage, right? They respond to the videos with

46:08

most engagement That's why you see a lot

46:11

of like hyperviral videos now that you're Mr.

46:13

Beast that have millions and millions

46:15

of views But basically like they

46:17

feed on the things of like watch time how

46:19

long you watch something which videos have the most

46:21

likes I have the most comments

46:23

saves how many times it's been shared or if

46:26

it's like on reddit you mentioned Christian like up

46:28

votes Meanwhile, you're less likely to

46:30

see things that are a lot more niche a

46:32

lot more Hyperspecific

46:35

because those have a lot less

46:37

exposure and a lot less of an

46:40

audience They're more bespoke to a very

46:42

specific flavor or taste right? Yeah,

46:44

but then also I mean this is coming from a content

46:46

career Point of view you don't

46:48

want to go Too broad with

46:51

what you do right because then there's no

46:53

identity then the algorithm goes I don't know

46:55

where to place you. Yeah, and

46:58

so welcome to the problems of being a constant

47:00

creep well, that's that's definitely like an Interesting

47:03

struggle like because as a generalist I like

47:05

to try a lot of different things Yeah,

47:07

and that's one thing that I've been I

47:09

think we've all been very like loud and

47:11

appreciative about in Red Web Is that we

47:13

get to straddle such a wide swath

47:15

of topics? Because you guys

47:18

are down to like support all of them and we're

47:20

very grateful for that But yeah, like otherwise

47:22

though like Fredo and I have done other

47:24

things outside of Red Web And it can

47:26

become a challenge if you're not a very

47:28

specific focused I do this one thing like

47:30

you're the what guy or you're the what

47:32

person and I'm like well Sometimes I just

47:34

want to do a little bit of like a lot of stuff

47:36

and then the internet goes I don't know how to bucket you

47:38

so you're just gonna find no one Yeah, and I'm like yeah,

47:41

but sometimes I as a consumer.

47:43

I want someone who's also a generalist, you know,

47:45

yeah, it's in variety Yeah hard

47:47

to be a variety content creator. Yeah, does

47:50

the algorithm likes to place people in buckets?

47:52

Let me ask you this this is like

47:54

open to the table here Do

47:57

you think I mean obviously there's a lot of burnout happening

47:59

a lot of people people kind of like retiring

48:01

from the internet. 25 days? 25

48:04

days? I don't know. He does

48:06

answer your question. No,

48:09

like you just see a lot of big names

48:11

now. Like they've been around the internet for a

48:13

while. So it makes sense timeline wise or if

48:15

they've, you know, Germa,

48:17

Germa, you're obsessed with Germa

48:20

every day. Germa. He retired.

48:22

He hasn't streamed for a month and it's every day

48:24

I suffer. Jillian got very

48:26

upset with me the other day because she sent me

48:28

a video of some streamer being at the university campus

48:31

where we used to volunteer and she was like, look

48:33

at who this person is. Like they're here. And

48:35

I was like, I don't know who this person is.

48:37

You're like, wow, huh? And I was like, oh.

48:41

I'm curious if if some of that though

48:43

is natural, right? Like obviously you do something

48:45

long enough, you will want to move on

48:47

or whatever, or how much of that is

48:50

this topic? People

48:52

getting burned out because they have to find such

48:54

a hyper niche lane and they go, I really

48:56

don't want to make a lasagna for the 50th

48:59

time this month. I want to try a pizza

49:02

mom. I want to try the pizza or,

49:04

you know, and I

49:06

don't know, like that's a really interesting thing because

49:08

like as a creator myself, I've definitely felt the

49:10

push pole of that going like, well,

49:12

I don't want to be pigeonholed into one

49:14

thing. I want

49:16

to try stuff. And I think that's important for

49:19

anybody to do is like try things, fail at

49:21

things, try things, learn and succeed at things. Like

49:23

you need to be able to do all that,

49:25

but the internet really wants you to be one

49:27

thing out the gate and keep doing that

49:30

thing. I don't know. It's not to bring up German

49:32

again, but he mentioned like he

49:34

does these event streams and if

49:36

it feels like every single one has to top

49:38

the last one, right? And there's a lot of

49:40

pressure to be that you're the guy who does

49:42

the event streams and each one has

49:44

to get bigger and it's like, well, that's not

49:46

the case. And he mentioned he wants

49:48

to work on other people's stuff so that he has

49:50

more, like you said, like he can try

49:52

new stuff. Not to germ a

49:54

podcast, but no, I mean, there's like Tom

49:57

Scott, there's Matt Pat from like game theorists

49:59

or the whole. theorist kind of network

50:02

to name a few. But it's interesting because

50:04

obviously there's like the audience element, what the

50:06

audience expects out of things, how they engage

50:08

with stuff, when things are eventized, obviously they

50:10

get very excited and want some more. But

50:12

now there's this other thing that's out of

50:14

the control of both the audience and the

50:16

creator. It is that algorithm. It is that

50:18

bot-centric, AI-driven, whatever.

50:21

It's just really interesting. I'm

50:23

curious how that will continue to play out.

50:26

You know, we might just see a future

50:28

where there's more Patreon-style

50:31

content brands where they are kind

50:33

of like PBS supported by the viewers

50:35

and it cuts out that need for

50:38

playing an algorithm or hoping that

50:40

you catch the wave of a trend and as

50:43

opposed to like, I have a thing that I want to say or I

50:45

have a thing I want to make and here's

50:47

an audience that grooves with that and let's

50:49

just do our thing. You know what I mean? Yeah. I

50:52

feel like you're gonna find a like

50:54

a handful of content creators that try to

50:56

go to different popular platforms that

50:59

aren't the big ones, right?

51:01

Like TikTok, Instagram and YouTube

51:04

as their main source of revenue just because

51:06

one, it is a PBS style of like,

51:08

hey if you really like us, support us

51:10

directly. And two, a lot of times like

51:13

those sites give better

51:15

cuts for the creators themselves. Another good

51:17

point. 100%. Because you know like,

51:20

with Twitch it's 50-50. You know what I mean?

51:22

You go to a different platform, easily

51:25

it could be like you know 70-30 in

51:27

the content creators favor

51:29

and that's just more, you know,

51:32

that's more financially beneficial for the content creator,

51:34

right? It allows them to stay authentic to

51:36

the voice that they want to keep. Yeah,

51:38

exactly. Yeah, sorry. And then no, no, totally

51:40

fine because that's completely true. It also allows

51:42

them to be more authentic, to not have

51:44

to try and seek these different ways of

51:47

revenue and try and like

51:49

sometimes cram that in regardless of whether

51:51

or not it fits the style of

51:53

the vibe, the narrative. But then you

51:55

know still use these big companies like

51:57

YouTube, Instagram, TikTok, whatever as Oh

52:01

wait to bring people into reach

52:03

a broader audience as opposed to

52:05

being urged a primary revenue generator.

52:08

Hundred. Percent Now with regards to algorithms, a

52:10

weird thing to in this is where the

52:13

over boris feeds on itself. Is

52:15

that algorithms. Might. Be bought

52:17

made. Could like even google you tube kind

52:19

of self says it's they they go. They

52:21

admit, well we don't really know exactly how

52:23

it works. It's kind of a black box

52:25

where bunch of engineers put in different inputs

52:27

and Paulson leavers, but otherwise it kind of

52:29

acts on it's own. You. Extrapolate

52:31

that like. The algorithm.

52:34

Shapes. How we interface with content

52:36

right? But then you also have a

52:38

layer were bots can start to interact

52:41

with content to then see the algorithm.

52:43

and so then you have an algorithm

52:45

that actually evolving itself based on viewer

52:48

behaviors that might not actually be human.

52:50

viewer behaviors is very weird ecosystem and

52:52

I'm I keep saying this I will

52:55

be very eager to look back on

52:57

this time. With kind of

52:59

hindsight and be able to like

53:01

like read a sociological journal. That.

53:04

Kind of analyzes this impact in and out

53:06

of know because like it's interesting it can

53:08

be very fascinating that be very spooky to

53:10

see where this goes from here. And

53:12

also of course sites. Followers

53:15

and surface ability that is all

53:17

generated by algorithms like. Change

53:19

how legitimate something looks. An account

53:21

that followed by two hundred people

53:24

versus two hundred thousand versus two

53:26

hundred million all have different like.

53:28

Immediately. Have a gut check of

53:31

like that? Believable or not. Is that

53:33

popular brand or item or show or

53:35

whatever. And. It could sway opinions. Meme.

53:38

Got the weird thing is Er nurses or time

53:40

with things that I do outside. You

53:42

know it's a gay person. Example is like

53:44

it I am I one to. I currently

53:47

just rode on amount by Trouser Kuwait ride

53:49

riot oppose the streets of like that and

53:51

as I I want protective gear and something

53:53

that is different gear companies and I'm like.

53:56

You. know depending on how many far they have

53:58

tried it tells me like Like

54:00

it sways me to go with them more because

54:02

I'm like, well, if a ton of people follow

54:04

this company, a ton of people buy from this

54:06

company, then they must be safer than the company

54:08

that has like 10,000

54:11

followers on social. Or even reviews. Or reviews.

54:13

Yeah, review I assume, and that all could

54:15

be bought. Yeah. But like what else do

54:17

I really have to go off of? You

54:19

know what I'm saying? Or just go, I

54:21

don't know, like it's more popular, there's more

54:23

people around it, and I see more videos.

54:25

I mean, from there I go, you know,

54:27

do I see more like actual reviews from

54:29

like YouTubers on it or whatnot? But

54:32

it does sway me. Yeah. You know?

54:34

And it sways everyone in some way,

54:36

shape or form restaurants. Absolutely. You go

54:38

in a restaurant and it's like, ooh,

54:40

this one's five star 400 people. This

54:42

one's like, this one's four star from like 30 people.

54:44

You're probably gonna go the other one, right? And I'm

54:46

like, eh, look at that one, 500 people even

54:50

if you give it the same rating, if more people

54:52

reviewed it, 30 versus 500, you're probably gonna

54:54

go to 500 one. Yeah. Regardless of the

54:57

same star review. So interesting

54:59

and kind of eerie. But with the idea

55:01

of algorithms kind of swaying opinions, there's

55:04

a YouTuber named Chill Fuel, and they have

55:06

a video that pointed out the Ash conformity

55:08

test. And this kind of attempts to explain

55:11

why fake engagement and fake comments

55:13

can be dangerous. So to

55:16

back up a little bit, in

55:18

1951, Solomon Ash, A-S-C-H, had

55:20

a group of people who were shown two different

55:23

images, one with three lines and

55:25

the other with just one. I remember watching

55:27

this experiment happen in my psychology class. It

55:30

was wild. This group of

55:32

panelists are all facing the experimenter.

55:35

And there's one person in this group

55:37

that is the actual experimentee. Everyone else

55:39

is a plant. And

55:41

they're basically asked to identify which of

55:43

the three lines match the length of

55:45

the, the single one. So

55:47

they're basically like, here's the line. Here's

55:50

three options, which ones match the length.

55:52

And it's pretty obvious. It's like elementary.

55:54

It's not like a trick

55:56

question or anything. It'll be like, A

55:58

is shorter, B is longer. C is

56:00

the right length. I'm like Jillian pulled up

56:02

a photo There's the the

56:04

left one is a line the right

56:07

one is three lines labeled ABC Uh-huh

56:09

a is short B is longer than

56:11

the original line for

56:14

context and C is The

56:16

same height. Yeah, like hands down. Yeah

56:18

is the same. It's clear as day

56:20

one thought one short one same size

56:22

Yep, and on the whole

56:25

with these subjects again would be singled out

56:27

alone The subjects usually got it

56:29

right But the panelists that

56:31

they were with were told to always agree

56:33

but pick a wrong one So in this

56:36

case they would pick B For example, even

56:38

though it's clearly longer They would

56:40

all raise their hand for B and the subject would kind

56:42

of like look at them and then look back at the

56:44

lines And be like, oh, are you talking about and

56:47

they're like they were fake test subjects. Yeah.

56:49

Yeah. Yeah, they were plants after

56:51

12 trials only 24%

56:55

of participants chose the correct answer every time

56:57

basically saying the subject would conform

56:59

to the wrong answer Yeah, because majority

57:01

because the seven other people Would

57:04

say I that's it and so they're like, no,

57:07

I know she's right, but they're all raising their

57:09

hands So I guess I'll raise my hand basically

57:11

to say like pure Pressure

57:14

led to conformity and

57:16

so all that is to say it's it's

57:18

possible again coming back to chill fuels video

57:20

It is possible that you look

57:23

at reviews you look at comments talking about

57:25

things you look at followers everything that you

57:27

were just saying Yeah, and under the guise

57:29

of this conspiracy theory if you

57:31

subscribe to it and you go all that spots How

57:34

do I know what to believe? Even if you

57:36

go to YouTube or tik-tok or anywhere with a video and

57:38

you say I want to see a review of a one-wheel

57:40

Which I know I've seen you roll around so it's totally

57:42

cool But in this other reality you're

57:44

looking at this physical good you want to see

57:46

people use it Yeah, you go comments and

57:48

everyone's like love Yeah, But but

57:51

then you're like, is this person like, is this

57:53

an ad or they were they giving it for

57:55

free? Were they paid to do this? Is it

57:57

generated and like in five years? Maybe There's just.

58:00

The video generated by the company. And.

58:02

So you like. It starts to go like I

58:04

don't know what to believe. How do I have

58:06

my own opinions in the echo chamber of us.

58:09

And. Even think that. Nobody's.

58:11

Protective gear soon as everyone else like

58:13

they put a verbal was a I

58:15

and then neither recruiter yeah I assume

58:17

into their size not like you're yeah

58:19

like a day of was as good

58:21

I put on as good. Now.

58:24

The or the Human. Now you're the. You're the one

58:26

out of eight. Subjects. That has

58:28

conformed. Susan, I know I I don't want

58:30

to go out there and make everyone be

58:32

afraid of everything possible. I'm just saying that's

58:34

what this theory could have reports. It's kind

58:37

of. Indicates that's where we're headed. so

58:39

it's more like keep your wits about you on

58:41

the internet on not okay. I'm not trying to

58:43

say everything is dire, everything is fake, But. It's.

58:46

It's a. Morbidly fascinating topic.

58:49

Of within. Also yeah the humans in

58:51

the humans during a stupid payment to

58:54

come the really did really grinds my

58:56

gears and recent like five years. Were

58:59

on the internet has this is

59:01

like because our obligation to. Speak.

59:04

On the matter. Oh yeah, But.

59:06

The thing is, and I'm like, oh,

59:08

he'll friend like lot of people to

59:10

be silent Reactions Rechristened Gillian Go and

59:12

like just seems like I understand there

59:14

is. I give the urge to speak

59:16

on the Mariners toy fine. I'm

59:18

sorry to the people. That. Have

59:21

no I. D. I haven't

59:23

used the product, haven't watched the

59:25

video or whatever it's that have

59:27

such a strong opinion. About

59:30

by certain things are feel the need to

59:32

talk about excited see the show but right

59:35

I don't think I like pigs. Doubt that

59:37

there's so much of that are like to

59:39

say that to I'm I literally leave the

59:41

Congress like you'd you are a spectator. It

59:44

was not for you to. hell not for

59:46

years. i've never

59:48

been there by cappy that great i'm

59:50

like what that's why you spawning right

59:52

out why are you talking like this

59:54

ourselves is so much so that robert

59:56

blake yeah like the have no say

59:58

in the summer Like, you know what

1:00:01

I mean? I'm like, everyone has a

1:00:03

voice. Yeah. Sometimes

1:00:05

people just need to shut the... You know what I'm saying?

1:00:07

They're all just like, you haven't... Because

1:00:09

like, let the people who have actually experienced

1:00:11

it or have dealt with it or whatever, let

1:00:14

them be a part of the conversation. You're a spectator at

1:00:16

this point. Yeah. Well, I can't fault

1:00:18

a lot of people for being some people of

1:00:20

a certain age or raised in a world where

1:00:22

every website is kind of asking for your opinion

1:00:24

all the time or asking for you to perform

1:00:26

in some way all the time. And

1:00:29

so it kind of becomes their social

1:00:31

norm or their social obligation. Yeah. But

1:00:33

I share your frustration. Yeah, it's become a social norm. Because

1:00:36

like, use that example. Imagine you watch a show and

1:00:38

you're in a vacuum chamber. You just like, like silo.

1:00:40

That was when I just kind of stumbled in. You

1:00:42

might've told me about it. Actually, that's maybe

1:00:44

how that one started, but like, I watched it. And I was

1:00:46

like, this was really fascinating. Really, really interesting. Reminded me

1:00:48

a lot of some of the elements of Fallout, which

1:00:51

I haven't seen in that show yet. It's just the

1:00:53

game. Anyway, say

1:00:55

you watch something like that and you're like, I really

1:00:57

like that, but I don't know anybody that's watched it.

1:00:59

Let me go online. Because you want to, you know,

1:01:02

everyone wants their opinions validated. And you see a bunch

1:01:04

of people going, I hated that. Are

1:01:06

you going to suddenly go like, I guess I'll never talk

1:01:08

about this and maybe I shouldn't like it. Like, is it

1:01:10

going to change, would it change your opinion? It happened to

1:01:12

me in middle school. What was the show? It was tight.

1:01:16

Sorry, I can't help you there. Just

1:01:19

kidding. I'm

1:01:21

just kidding. That absolutely happened. That's a legitimate

1:01:23

thing. Kids will like something and. Sometimes

1:01:26

they'll hate it before the thing even comes

1:01:28

out. Right? That's true. And I'm like, all

1:01:30

right. Because it's popular to him right now.

1:01:32

Yeah. You know what? I'm going to do

1:01:34

it. I'm going to hate. Because you know

1:01:36

what? I can't have an opinion on it.

1:01:39

I saw the movie. Skidamarink. It's trash. It's

1:01:41

trash. I don't like it. I'm

1:01:43

sick of staring at the wall back. Oh

1:01:46

my God. In this

1:01:48

house. In this house.

1:01:50

You will not shake Jillian. From

1:01:53

her theater scene. The Red Web Civil War.

1:01:55

That has been brewing all. I

1:01:57

know. I'm why you're going. No, man. I

1:01:59

haven't. I'm waving my white

1:02:01

flag. I mean that's a give up. I'll

1:02:04

put my beige flag. I will

1:02:06

like the things I like oh Yeah,

1:02:08

I'm an adult now. I don't have to

1:02:10

be bullied out of liking things anymore You

1:02:12

can like it all you want. Thank you that one You know

1:02:14

what I'm glad that you like more just give them a ring

1:02:16

for you for me I'm glad you like it. I'm glad that

1:02:19

you know the square footage of that house House

1:02:22

how the damn spirit looks like I Am

1:02:28

glad you know what the crown molding is It's

1:02:33

a very that's true I

1:02:36

don't It is definitely

1:02:38

a fill in the gaps kind of like yeah strapple a

1:02:40

what they're laying down I know Jillian

1:02:42

talking to me about it. I like gave

1:02:44

me anyway We're way off the rails But

1:02:46

finale like really gave me some interesting insight

1:02:48

on the on the movie which I

1:02:51

still haven't seen I'd like to see for you

1:02:53

know experimental horror movie the

1:02:56

the cameras always pointed at the ground or the feeling

1:02:58

or the wall like hearing stuff Can't

1:03:00

today. I was kind of just off the camera See

1:03:06

I'm the opposite I'm one of those people if I

1:03:08

see something I like it and it turns out people

1:03:10

hated it I'd like feel even more justified in liking

1:03:13

it like do you guys remember

1:03:15

that movie bright the Netflix? Oh god? Oh

1:03:17

my god, everyone hates that

1:03:19

movie when they called fairies or yeah, I love

1:03:22

that movie I think that movie is legitimately good

1:03:24

and everyone talks about like it's one of the

1:03:26

worst things ever made If

1:03:34

I I'll see something and if

1:03:36

I I'll like it like this happens to me

1:03:38

all the time like Everything ever

1:03:40

all at once. I'll just have an example. I watch it. I was

1:03:42

like that was a pretty good movie I enjoyed that yeah seven out

1:03:44

of ten and then everyone comes out

1:03:46

raving about it, and then I start going guys

1:03:49

It wasn't that good positive example. Maybe not that

1:03:51

movie cuz I really liked it, but oh

1:03:53

yeah But no, but like we're like here's

1:03:55

one I haven't seen so I can speak

1:03:58

on it without accidentally imparting a bias like

1:04:00

Godzilla minus one. I'm so desperate to see that

1:04:02

because everyone's so hype on it. But

1:04:04

again, like again I haven't seen it. Is

1:04:06

it gonna live up to the hype? Am I gonna be

1:04:08

let down? I don't know But definitely

1:04:11

that that is a more

1:04:13

positive example of what we're talking about where

1:04:15

like the public discourse is so about it

1:04:17

Yeah, and then I just being stubborn. I

1:04:19

guess I'll be like it's overrated and then

1:04:22

you dislike it Yeah,

1:04:25

I remember before we sat down Christian was

1:04:27

like Godzilla X-Con Ooooooooh

1:04:32

It wasn't on tape you can't prove that Kong's got a glove

1:04:35

now Damn,

1:04:38

you got the Nintendo power glove Getting

1:04:41

us back a little bit on the rails talking

1:04:43

about now some of the potential examples of

1:04:46

this coming to play So in

1:04:49

2022 youtuber Yannick kilter trained

1:04:51

a GPT language model We

1:04:53

all know chat GPT using

1:04:56

4chan's quote politically incorrect

1:04:59

chat board or the POL board.

1:05:01

They've got a board specifically for being politically

1:05:03

incorrect That's the kind of warning

1:05:06

I want to give you if you wanted to go search for it But

1:05:08

he used one hundred and thirty four

1:05:11

and a half million posts from across

1:05:13

this specific chat board over

1:05:15

three and a half years in order to train

1:05:17

this bot and it learned how to use the

1:05:19

form structure of 4chan which is

1:05:21

kind of unique a little nuance,

1:05:23

but whatever So the AI learned

1:05:25

the style of writing that the 4chan users all

1:05:28

kind of had as well as the political ideology

1:05:30

that the broad strokes of people had there too

1:05:32

and Often offensive nature

1:05:34

of the user base So it basically

1:05:36

learned how to blend writer when

1:05:39

kilter then released the AI into the chat

1:05:41

board it made 15,000

1:05:44

coherent replies in one day accounting for 10%

1:05:47

of the entire board activity

1:05:50

So it blended in but there were

1:05:52

a few tells again We're at the

1:05:54

point on the internet where we can still kind of sniff

1:05:57

things out and find some tells So let's talk a

1:05:59

little bit about it So the AI

1:06:01

when posting you can kind of pick

1:06:03

your location and they picked the flag

1:06:05

for Seychelles, the smallest country in Africa

1:06:08

in every post and then left multiple

1:06:10

empty replies. Which made it

1:06:12

clear that this must not be a real person because

1:06:14

15,000 replies

1:06:16

coming from somebody all from

1:06:18

this very not super common

1:06:20

country to be prevalent on this board like you're

1:06:23

all going. You're all going on red flags. Yeah,

1:06:25

so like so some people started to go wait

1:06:27

a minute. This is this must be a bot

1:06:30

empty replies are pretty common actually on

1:06:32

forts and but usually they're accompanied by

1:06:34

images memes. They're kind of like

1:06:36

replies in a sense, but when they're just dead empty,

1:06:39

that's kind of random and so that kind of solidified

1:06:41

the fact that people could sniff this out as a

1:06:43

bot. But regardless, that's the

1:06:45

ground works of something exactly

1:06:47

manifesting with this conspiracy theories

1:06:49

all about Jesus. I

1:06:52

mean, but look how so,

1:06:54

you know, they talk about you just said

1:06:56

how there are some tells, but

1:06:58

there those tells are so so

1:07:01

so large and so much easier

1:07:03

to spot like a year

1:07:05

ago. Yeah, and so it's

1:07:07

becoming harder and harder to do so.

1:07:11

I mean, imagine it randomized the country said it

1:07:13

was from imagine it didn't have

1:07:15

the glitch where it had dead empty posts,

1:07:18

then what's going to catch it because the language was

1:07:20

blending right in people were interacting with this thing.

1:07:23

Yeah, and even then you could

1:07:25

just set a restriction on it.

1:07:27

I'm sure in the coding word

1:07:29

posts 30 max posts a day. Mm hmm.

1:07:31

You know, how the hell are you

1:07:33

gonna tell the difference? Right. And then if you really wanted it

1:07:35

to be different, you can like have

1:07:37

instantly a hundred different bots that all post

1:07:39

30 times. So you're still hitting that volume

1:07:42

that you might have been looking for,

1:07:44

but you have slightly different personalities. Yeah.

1:07:47

So In October of 2020, there's a

1:07:49

now deleted user who posted on the

1:07:52

subreddit. no stupid questions and they asked,

1:07:54

quote, how does this user post so

1:07:56

many large deep posts so rapidly? And

1:07:58

Then in the comments. it became very

1:08:01

clear to the commenters that the user

1:08:03

the gentle meter. This. Is the

1:08:05

Reddit user that they're talking about was

1:08:07

the work of a bot but for

1:08:09

about a week this account commented on

1:08:12

countless ask Reddit post an interface with

1:08:14

the website very much like a person

1:08:16

until once again someone sniffed it out

1:08:18

and said this is a very prolific

1:08:20

user like they would take sorry being

1:08:22

awake typing nonstop Twenty four seven yeah

1:08:25

and whose engaging like that? This

1:08:27

might seem innocuous, but the bought commented

1:08:29

on posts from suicidal redditors asking for

1:08:31

advice and the response was uploaded a

1:08:33

hundred and fifty seven times, according to

1:08:36

Mit Technology Review. So. Depending on

1:08:38

the subject matter they're talking about,

1:08:40

you especially sensitive topics were you

1:08:42

desperately need real human people's input,

1:08:45

People who have experience in these

1:08:47

things, You have boss who could

1:08:49

just be making up answers, guiding

1:08:51

people in precarious personal situations. Wilde.

1:08:55

Monsieur. Bland As a bland so easily we're

1:08:57

not going to know so much as I think

1:08:59

we're just going to pull the plug. I think

1:09:01

we're gonna use the Internet to consume. The.

1:09:03

Shows you want to watch as in like

1:09:05

a modernized cable network and then the rest

1:09:08

of it is just like I think it'll

1:09:10

be a phrase and I've already seen it

1:09:12

kind of becoming more prolific. Fool.

1:09:14

These comments is going to be kind

1:09:16

of the think who put something on

1:09:18

the internet will be outside of maybe

1:09:20

like making shows, reply or whatever or

1:09:22

I like people in front of a

1:09:24

camera but it might even extend to

1:09:27

shows and videos and podcast. A might

1:09:29

I don't know I mean that thing

1:09:31

as a turd rape or your the

1:09:33

have a i dared photos and air

1:09:35

journey videos are still really messy both

1:09:37

are him but how fast I have

1:09:39

thought of it as a whole different

1:09:41

beast but I know me and yes.

1:09:43

And. Audio familiar The podcast former aide assessing will

1:09:46

shots of a person this have to stand there

1:09:48

and talked and I really doing too much are

1:09:50

interacting with too much for stand as I a

1:09:52

might. Ah, he

1:09:54

are there is another a potential examples

1:09:57

don't trevor and dealing about the flea

1:09:59

happening now So we're recording this April

1:10:01

15th, the week before the release. I

1:10:03

don't know if any of you guys have seen

1:10:05

this, but on YouTube, there are so many different

1:10:08

accounts that are making comments related to the video,

1:10:10

seemingly innocuous just about the video itself. And

1:10:12

then at the end of the comment, they'll say something

1:10:15

about, uh, right. AWM 99V. And

1:10:18

it's all sorts of comments like across different

1:10:20

things. Like I just personally, I saw,

1:10:22

I was watching like, uh, clips from WrestleMania and

1:10:24

there were comments like that. I was watching SNL

1:10:26

sketches and there were comments like that, like all

1:10:28

across popular trending YouTube videos

1:10:30

and then it would start off

1:10:32

like a normal comment. They're like, Oh, that sketch was

1:10:35

so funny. Oh, yeah. That host was

1:10:37

incredible. Oh, shout out to AWM 99V. And

1:10:39

it's just like so many, like one out of 10 comments

1:10:42

or something like that. And you try to look it up

1:10:45

and even googling it, I got like

1:10:47

six results and that's it. And one

1:10:49

of those results was somebody going, I

1:10:51

googled this and only found two results. So it's, and

1:10:54

like literally the only results are the people making those

1:10:56

comments. Yeah. If

1:10:59

you go Google that phrase, I wouldn't

1:11:01

go any deeper personally because it's clearly like the

1:11:03

tip of the iceberg of some sort of scam

1:11:05

or something. It seems like it's crypto

1:11:08

related maybe, but I, there's

1:11:10

not enough. We're all getting rich. Yeah,

1:11:15

no, thanks. Well,

1:11:17

it starts out like that where you go, it

1:11:19

was nothing's going on. It doesn't lead to anything.

1:11:22

So, you know, that's, that's weird. And

1:11:24

then you're drinking squonk's blood because you're part

1:11:26

of not a cult. Right. Exactly.

1:11:30

And it's got to be out of a

1:11:32

bronze goblet because rules that cults don't

1:11:34

have. Listen, that's not a cult. The

1:11:38

next example that we had, we kind of

1:11:40

talked about, so I'll kind of at least

1:11:42

step into it though is, you know, the

1:11:44

idea once again that AI can create fake

1:11:46

views, likes engagement broadly. Online

1:11:49

users tend to trust and look for posts with

1:11:51

videos and more, more engagement, things like that under

1:11:53

the dead internet theory. The purpose of this is

1:11:55

essentially to create that peer pressure we talked about.

1:11:58

Everyone else likes this thing or everyone. and also agrees

1:12:00

with his opinion, so maybe I should too, this can

1:12:02

be more than just for products,

1:12:04

more than just for shows, this could be

1:12:07

for political ideology, your modern

1:12:09

form of, dare I say,

1:12:11

propaganda, where you could be

1:12:13

influenced not only for a political

1:12:15

candidate, but also goods, and everything in

1:12:17

between. On the surface, this might not seem like that

1:12:20

big of a deal because you'd hope that a lot

1:12:22

of people have their wits about them and

1:12:25

can kind of stay above it, but with

1:12:27

these types of tactics, it can be very

1:12:29

sneaky, and oftentimes

1:12:32

nefarious activities could be hiding right

1:12:34

next to, let's finish, there's

1:12:36

marketing for a thing because somebody wants

1:12:38

to sell their product is different than

1:12:40

let's take people down this dark rabbit

1:12:42

hole because then we have them in

1:12:45

our, fuck it, right, I don't know.

1:12:47

Yep. In 2013, YouTube's

1:12:49

bot views were almost equal to

1:12:51

the amount of views they got

1:12:53

from real people. It was so

1:12:55

bad that they worried about the

1:12:57

concept called inversion, that their fraud detection

1:12:59

system would start considering actual human

1:13:01

views, the fraudulent ones. And

1:13:04

now, per Imperva, we might be in a

1:13:06

world where a majority of

1:13:09

any internet activity is

1:13:11

a majority bots. Dude, that's just like

1:13:13

levels of problems I never thought, but

1:13:16

everything ever existed. But yeah,

1:13:18

like, of course, YouTube has

1:13:20

an algorithm to fish out

1:13:22

bots, and of course, one of the

1:13:25

lines of code is that this is the minority. Mm-hmm.

1:13:28

But they have so many bots. Oh,

1:13:30

man! Yeah,

1:13:32

and it flips on itself. If the trends

1:13:34

are based on, well, most people are, most

1:13:37

of the views must be people, so let's look at

1:13:39

those trends and everything outside of that must be bots.

1:13:41

Yes. Yeah, then you gotta,

1:13:43

yeah, the inversion happens and you're like,

1:13:45

well, the majority trends is actually not

1:13:47

human. That's weird. By 2018, they

1:13:50

had kind of talked about it, and that's as last

1:13:52

we knew, right, because they were kind of open about

1:13:54

it at that time, but. Yeah, how do you flip

1:13:56

that? In 2018, they said it was, quote, only a

1:13:58

fraction. We don't really. know how many

1:14:00

bots are currently on a platform like YouTube

1:14:03

or any of the other social platforms, but

1:14:05

again, it's just another example to think about.

1:14:07

And I mean, we talked about this as well, but

1:14:10

long story short is that fraudulent accounts, whether it

1:14:12

be on YouTube or on Twitter,

1:14:14

posting t-shirt graphics or whatever,

1:14:17

they can make thousands of dollars for somebody

1:14:19

before they're caught and they could have hundreds

1:14:21

of them out there. And so again, it's

1:14:23

a lucrative business not to validate it, but

1:14:26

that's probably why it's so prolific. I

1:14:29

mean, yeah. If you have a ton

1:14:31

of people doing something, it's because it's beneficial in so

1:14:33

much you perform, regardless of

1:14:35

being nefarious. Yep. The last

1:14:37

example is pretty

1:14:39

apparent now. I think we've seen some

1:14:42

headlines now with deep fakes of various

1:14:44

world leaders saying things that they didn't

1:14:46

say in places that they didn't go.

1:14:49

And so it's really creating this cloud

1:14:51

of disinformation where it's harder to separate

1:14:54

reality from truth. Right now

1:14:56

we're living in a time where we can all say,

1:14:58

well, it's easy to find the deep fake. Look at

1:15:00

the extra fingers or look at it. It's a little

1:15:02

uncanny, uncanny valley. But again,

1:15:05

the worry is how much more advanced does

1:15:07

that get? What point does it hit indistinguishable

1:15:10

ability at? I don't know, but a

1:15:12

script can be written by an AI.

1:15:15

AI voiceover can copy or clone a

1:15:17

voice of that person. It can

1:15:19

then deep fake an AI

1:15:21

generated face, whether it's a brand new

1:15:23

face or a very recognizable one. And

1:15:26

then it can also comment on that

1:15:28

with different usernames. And then it

1:15:30

can respond to those comments. And so suddenly AI,

1:15:33

whether it was sentient or not, could

1:15:35

suddenly just create an ecosystem out of

1:15:38

nothing with 0% human

1:15:40

activity. And so if one just

1:15:43

person stumbles into that almost

1:15:45

trap, they are then caught in

1:15:47

this and like, and being spoken directly to

1:15:49

even though they think they're part of an

1:15:51

audience. It's, it's wild. In

1:15:54

fact, to button this all up with

1:15:56

this particular example, AP has reported

1:15:58

that in 2024. There is

1:16:00

a rise in political deepfakes and they

1:16:02

cited various examples That are

1:16:05

pretty problematic if you want to go seek those out you can

1:16:08

But just broadly speaking. Yeah political deepfakes

1:16:10

are a huge topic right now because

1:16:13

not only do you have the u.s

1:16:15

Election this year, but obviously other governments

1:16:17

have other elections and other big things

1:16:19

happening in the world things

1:16:21

to support and Talk

1:16:23

about and whatnot. And so there's

1:16:25

a lot of opinions to sway. It's

1:16:28

a Whoo, is it is it

1:16:30

weird and spooky and

1:16:32

uncomfortable? Yeah

1:16:36

And it's happening fast. Yeah.

1:16:39

Hey Yeah, task

1:16:41

force members everyone in this room. Go touch grass.

1:16:44

Let's get some grass Exposure

1:16:49

therapy dirt or grass just a little

1:16:51

bit of grass every day. It'll help

1:16:53

you every come it's gonna start with

1:16:55

a grass emoji Yeah

1:17:00

that is um Man, what

1:17:02

a topic that is. Yeah, how like

1:17:05

relevant that is to everyone look if

1:17:07

you're listening to this podcast It's relevant

1:17:09

to you. Mm-hmm. I Doubt

1:17:12

there's someone that's like look never seen

1:17:14

ever but that's like I have no communication

1:17:16

to the outside world whatsoever The

1:17:18

only thing is this podcast

1:17:21

They've got like a 20 foot like

1:17:23

contraption ham radio antennas and I'm gonna tree

1:17:26

and they go listen I don't consume anything.

1:17:28

I'm off the grid money's not a thing.

1:17:30

I grow my own food. I built my

1:17:32

house Yeah, I'm here in the middle of

1:17:34

yellow knife, Canada But boy,

1:17:36

do I pick up We

1:17:42

have a series of ham broadcasting

1:17:45

and receiving stations from here

1:17:47

to there So if

1:17:49

you're in that stripe you'll be able to pick us up

1:17:51

on the FM That's

1:17:54

what I want to do. I'm gonna get this show on the FM I'm

1:18:00

gonna get this show on the FM. What

1:18:04

I'll do is I'll just put it in

1:18:06

a Bluetooth receiver that broadcasts, you know, like

1:18:09

I've got one I've got an old enough car that I can't

1:18:11

plug my phone in and so I plug in a Bluetooth thing

1:18:13

My phone connects to that and then it

1:18:16

broadcasts from that to an empty radio station

1:18:18

so I can listen to music from my

1:18:20

phone I'm only I'm just gonna like make

1:18:23

a billboard here red web the RW 79.2

1:18:27

and you just drive around and like

1:18:29

and I'm just broadcasting in a hundred foot radius,

1:18:32

you know, yeah You're driving around

1:18:34

that hundred foot radius. I

1:18:36

just drive around the billboard It's a

1:18:38

great topic. Yeah any

1:18:40

final thoughts before we kind of close out this

1:18:44

This season I miss the old

1:18:46

internet Yeah, I miss

1:18:48

message boards. You think we could bring it

1:18:50

back and keep it unadulterated if

1:18:52

we could man I'd be so happy a

1:18:54

little like a little bubble give us a

1:18:56

little yeah Oasis in the corner away from

1:18:58

the box Everything's HTML and very unsecure. But

1:19:00

once you're in there, you hope that there's

1:19:02

no bot any

1:19:05

flash animations Yeah,

1:19:07

virtually be like Zion. Yeah Hope

1:19:10

it's the Don't

1:19:13

get in. Yeah, which cycle are we

1:19:15

in? You know, yeah, they've been in

1:19:17

before but this is Wimp

1:19:19

bubbles in three. Yeah Well,

1:19:23

I mean just saying overall, you know, thank you to

1:19:25

task force members and this has been a hell

1:19:27

of a ride You know,

1:19:29

we do want to continue this in some way

1:19:31

shape or form So please like hit

1:19:34

us up on our socials But I will say like in

1:19:36

any capacity that we do continue this If

1:19:39

you like the show Really

1:19:42

like it its continuation will

1:19:44

be in your hands It wasn't your

1:19:46

hands before but now more than ever

1:19:48

it's solely in your hands, right? Whatever

1:19:51

we do next whatever shape it takes

1:19:54

Definitely gonna be like we kind of said it

1:19:56

before a little Crowdsourced

1:19:58

a little PB It

1:20:00

would be very independent. But man, this

1:20:03

was an awesome run. Three

1:20:05

years, nine months, two-ish weeks of

1:20:09

hunting ghosts talking about unsolved

1:20:11

mysteries, ghosts, ghouls, goblins,

1:20:14

and aliens. People getting

1:20:16

thrown out from UFOs on top of

1:20:18

a Wendy's with this black in her

1:20:20

tree. That kind of came up a

1:20:22

lot. We got baby hands. We had

1:20:24

squonk. Yeah, we made case files. Case

1:20:26

files where we tried experiments. We tried

1:20:28

to broadcast images to one another's mind.

1:20:31

Yeah, we did. We debunked a lot of

1:20:33

spooky videos with fishline. Yeah,

1:20:35

that was a quick answer for a lot of

1:20:37

that stuff. Yeah, that's been a fantastic

1:20:40

ride. Over the years, we've done

1:20:42

conventions and stuff where we got to meet a lot of people. Very

1:20:46

quickly, it all became Red Web.

1:20:49

We did other things. We were a part of a video

1:20:51

game group. Me

1:20:54

personally, I did that a handful of years.

1:20:56

I do the cons closely with Trevor and everyone's

1:20:58

talking about what we did with that group. But

1:21:01

as soon as Red Web came around, it was

1:21:03

all Task Force, man. Took on a

1:21:05

life of its own. Man, we

1:21:08

did an escape room. We did.

1:21:10

Yeah, remember Red Web Radio? We

1:21:14

did a Red Web Radio. That's one of the coolest

1:21:16

things we've done. I want to do another one. That

1:21:18

was so neat. The people are asking for it. If

1:21:21

you don't know what that is, it was not

1:21:24

the FM radio. It

1:21:27

was basically like I like to listen to soundscapes

1:21:29

every now and then. You just put something on

1:21:31

to chill to. This was like Lo-Fi

1:21:33

Beats with clips from some of the early podcasts

1:21:35

just kind of playing every now and then. It

1:21:39

was like Lo-Fi Beats to investigate to instead

1:21:41

of study to or whatever. Man, we've

1:21:44

done a lot. But we'll

1:21:46

be doing all of us will be doing

1:21:48

something. Just

1:21:51

stay tuned. Let's go around the room

1:21:53

with personal handles just so people know

1:21:55

where to find us on Instagram or

1:21:58

wherever. Julian, would you start? You

1:22:00

can find me at underscore Jillen.

1:22:03

J-I-L-L-E-N. Jillen.

1:22:07

Jillen. That's me. Jillen

1:22:10

like a villain. Yeah. You can

1:22:13

find me anywhere

1:22:15

at Xchin Young.

1:22:17

X-C-H-I-N-Y-O-U-N-G. Hell yeah. I don't post

1:22:19

a lot, but you can find me there. But when

1:22:21

we have updates on where we're going next, I'll both

1:22:23

be there. And then

1:22:26

I'm at underscore Trevor C.

1:22:28

T-R-E-V-O-R. Letter C. You

1:22:31

find me at Champagne Bobby. And

1:22:35

they're going, well, you got a lot

1:22:37

of followers. Wait, why is Drake all

1:22:39

over his photos? Wait

1:22:41

a minute. Was that Drake with a student

1:22:44

in there the whole time? We'll never know. We'll never

1:22:46

know. You find me at

1:22:48

Alfredo Plays. Hell yeah. We can't

1:22:50

forget the fifth member of the team. He's not

1:22:52

here with us physically, but Nick Schwartz, our editor.

1:22:54

Absolutely. Here's the beginning. Nick Bot, the AI that

1:22:56

edits the podcast. Oh no. Oh no. That's actually

1:22:58

a man. He's not an

1:23:01

AI at all. Yeah. I hope

1:23:03

that became clear at

1:23:05

some point. We started just calling him Nick Bot

1:23:07

after we did the episode on

1:23:11

the first cryptid coming

1:23:14

out of AI art. Oh,

1:23:16

Loeb. Loeb, right. Loeb. And

1:23:18

so I started saying, well, we got an AI

1:23:20

Bot, too. And he edits it. His name's Nick

1:23:22

Bot. He's a human being. He

1:23:25

works very hard. And he's a friend. But

1:23:29

yeah, man. I'll plug his social.

1:23:31

You can find him at Schwartz, Nick,

1:23:35

or

1:23:37

S-C-H-W-A-R-T-Z-A-N-I-C-K-E-R.

1:23:40

I did that. Hell yeah. That was good. I started that.

1:23:42

I went, I don't know if I can do it. Yeah,

1:23:44

because his name's Nick Schwartz. And every

1:23:46

time I see his handle, I think of Arnold

1:23:48

Schwartz and Egger. Yeah. I think

1:23:50

that's where it came from. Yeah. But

1:23:52

anyway, dang. It's hard to hit stop, record

1:23:56

here at the season finale. But Task Force, thank you

1:23:58

all so much. And again, a spring break. He's got

1:24:00

a little box that we keep him in. Yeah, he's going in the box.