0:00

Calling all Call of Duty

0:02

fans, the iconic map Verdansk

0:04

returns to Call of Duty

0:07

war zone. Starting April 3rd,

0:09

you'll be able to drop

0:11

back into Verdansk. Experience all

0:14

the chaos and relive the

0:16

thrill you've been. Missing. Not

0:18

only will you get the

0:21

classic Battle Royale experience we

0:23

all know and love, but

0:26

Verdansk is back with gameplay

0:28

updates and the return of

0:30

Verdansk era weaponry. That's right.

0:33

You'll experience Verdansk like never

0:35

before. Smoother movement. stunning visuals

0:37

and new mechanics. Whether you're

0:40

dropping in solo or teaming

0:42

up with your squad, it's

0:44

time to come home to

0:47

Verdansk. So download Call of

0:49

Duty Warzone for free and

0:52

drop in a Verdansk on

0:54

April 3rd. Read it in

0:56

for mature. I want to

0:59

let you know we've restocked

1:01

all your favorites on the

1:03

merch site. Everything is in

1:06

stock. You can show up

1:08

the Yovan store.com and thank

1:10

you so much for the

1:13

support I have some new

1:15

tour dates to tell you

1:18

about I'll be in Chicago

1:20

Illinois on April 24th at

1:22

the Win Trust Arena Fort

1:25

Wayne Indiana on April 26th

1:27

at the Allen County War

1:29

Memorial Coliseum and Miami Florida

1:32

on May 10th at the

1:34

Kasea Center Get your tickets

1:36

early starting Wednesday, February 19th

1:39

at 10 a.m. with pre-sale

1:41

code, Rat King. We also

1:44

have tickets remaining in East

1:46

Lansing, Victoria, B.C. in the

1:48

Canada, College Station, Belton, Texas,

1:51

Oxford, Mississippi, Tuscaloosa, Alabama, Nashville,

1:53

Tennessee, Winnipeg, in the Canada

1:55

and Calgary, in the Canada.

1:58

All tickets at theovon.com/T-O-U-O-U-L- Today's

2:00

guest is from Los Alamos,

2:02

New Mexico. He's a leader

2:04

in the world of business

2:07

and technology. He's an entrepreneur.

2:09

He started Scale AI. A

2:11

company recently valued at

2:13

$14 billion. He started

2:16

it when he was only 19.

2:18

And at 24, he became the

2:20

youngest self-made billionaire in the

2:23

world. We talk about his company,

2:25

The Future of AI, and the

2:27

role it plays in our human

2:29

existence. This was supereducational for me.

2:31

I hope it is for you.

2:33

I'm grateful for his time. Today's

2:36

guest is Mr. Alexander Wang.

2:54

Alexander Wang man, thanks

2:56

for hanging out, bro. Yeah, thanks

2:58

for having me. Yeah, good to

3:00

see you, dude. Last time was at

3:02

the inauguration. Yeah, what'd you think of

3:05

that? Like, what were your thoughts after

3:07

you left? Because you and I ended

3:09

up, we like left out of there

3:12

and then we got lunch together, which

3:14

was kind of crazy. It was, there

3:16

was a, were you there the whole

3:19

weekend? No, I just got there, the

3:21

day, the morning of the inauguration, the

3:23

new administration is really excited about it

3:25

and wants to make sure that we

3:28

get it right as a country. So that was

3:30

all great, but it was kind of a crazy,

3:32

the whole like event, like everything was pretty

3:34

crazy. I don't know, what'd you think? I

3:36

mean, when I saw Connor McGregor show up,

3:38

that's when I was like, shit is, where

3:40

are we? It felt like the boost bizarre.

3:43

I mean, you were there Sam Altman was

3:45

there. I was just like, it was like,

3:47

what happened here? And part of that's because

3:49

Alex Broustowitz, you know, totally, yeah. I mean,

3:51

it's all because of him that we all

3:54

went. But just the fact that he would

3:56

bring these people together, it kind of makes

3:58

you question. Crossover, it's like a. It's like

4:00

in those like TV shows when you

4:02

have those crossover episodes. Yeah. Oh yeah,

4:05

when like the Harlem Globe Trotters and

4:07

it would be like them versus like

4:09

the care bears or whatever would show

4:11

up. Exactly. Yeah, this seems. Yeah, some

4:13

cross-pollination. Yeah, what'd you think when Connor

4:16

showed up? Was that strange to see

4:18

him? Like was there somebody you thought

4:20

that was strange, like unique to see

4:22

there for you? Well. Yeah, I mean

4:24

you, Connor, like the Pauls, I don't

4:27

know, the whole, the whole thing was

4:29

pretty crazy. I see Sam all the

4:31

time and I see, I see some

4:33

of the tech people all the time.

4:35

I mean, it was funny crossover and

4:38

it was obviously like, so many people

4:40

flew in to be able to go

4:42

to like the outdoor inauguration, right? And

4:44

so, I mean, there were so many

4:46

people in the city, we ran into

4:49

them obviously, but like there were so

4:51

many people in DC, in DC, who

4:53

were just around for the inauguration for

4:55

the. Yeah I didn't even think about

4:57

that all there was a couple hundred

5:00

thousand people probably that it's got kind

5:02

of displaced in a way and suddenly

5:04

there's in bars or whatever just like

5:06

buying like there was like people I

5:08

saw walking just adult onesies and shit

5:11

that said trump or on the best

5:13

crazy shit yeah um Yeah, thanks for

5:15

coming in man. I'm excited to learn

5:17

about AI. I know that that's a

5:19

world that you're in. You're in the

5:22

tech universe. And you're from New Mexico,

5:24

right? Yeah. And so when you grew

5:26

up there, did you have like a

5:28

strong sense of like science and technology?

5:30

Was that like part of your world?

5:33

Like did your parents like lead you

5:35

into that? Like what was just some

5:37

of your youth like there? Did you

5:39

watch Oppenheimer? Yeah. Yeah. So the so

5:41

all the New Mexico shots in Oppenheimer,

5:44

that's exactly where I grew up. Oh,

5:46

damn. So it was like originally where

5:48

all the scientists came together and and

5:50

did all the work in the atomic

5:52

bomb. And there's still this huge lab.

5:55

That's basically the everybody I knew effectively

5:57

like their parents worked at the lab

5:59

or were some affiliate with the lab.

6:01

like Nookville over there, huh? Nookville, yeah.

6:03

Is it scary like that? It was,

6:06

you know, at a level of mystery,

6:08

like is your prom like last night

6:10

under the stars or something like? There's

6:12

this one museum, well the funny thing

6:14

is like, first you hear, you hear

6:17

this, you learn the story of. the

6:19

atomic bomb and the Manhattan project, like

6:21

basically every year growing up, because there's

6:23

always like a day or a few

6:26

days where, you know, there's a substitute

6:28

teacher and they just, they just play

6:30

the videos. So you're just like, yeah,

6:32

an alcoholic or something. Yeah, yeah. And

6:34

recreational use or use that term. That's

6:37

crazy that that just like every year

6:39

you guys have like, yeah, just like

6:41

blast it Thursdays or whatever. And it's

6:43

just, yeah, you're just learning, you're learning

6:45

again about the Manhattan project. there's a

6:48

little there's a little museum in town

6:50

which is like you walk through and

6:52

there's like a life-size replica of the

6:54

nukes so it's it's pretty wild yeah

6:56

and where did they drop the atomic

6:59

bombs on they dropped them on Asia

7:01

right Hiroshima they yeah they dropped them

7:03

on in Japan yeah Hiroshima Nagasaki is

7:05

it crazy being like semi Asia part

7:07

Asian Yeah, my parents are Chinese. Oh

7:10

nice man. Is it crazy being Asian

7:12

and then having that happen to Asian

7:14

people with a bond like? Is that

7:16

a weird thing there or it's nothing?

7:18

No I don't think so. I mean

7:21

I think the thing is like, you

7:23

know, so there weren't, I didn't grow

7:25

up with very many Asians, because in

7:27

that town it was, you know, it's

7:29

in New Mexico, there's very few Asians

7:32

in New Mexico. So. I was one

7:34

of the only Asian kids in my

7:36

class growing up and so I didn't

7:38

think that much about it honestly. But

7:40

then like, but it is super weird,

7:43

you know, you grow up and you

7:45

learn about this very advanced technology that

7:47

had this like really, really big impact

7:49

on the world. And I think that

7:51

shaped. Yeah, it's like the scientific John

7:54

Jones over there really. He's New Mexican,

7:56

isn't he? Lives in Albuquerque, yeah. He

7:58

does? I think he does, yeah. Oh,

8:00

sweet man. Yeah, yeah. Oh, yeah, so

8:02

you, so you're there, there's this, there's

8:05

this, there's this. Energy always there of

8:07

this creation, so probably the possibility of

8:09

creation maybe was always in the air.

8:11

I'm just wondering like how did you

8:13

get formed kind of, you know, like

8:16

what's your origin story kind of. It

8:18

was super scientific because, you know, there

8:20

were all these, there were all these

8:22

presentations around what were the new kinds

8:24

of science that were going on at

8:27

the lab. So there's all these chemistry

8:29

experiments and these different like earth

8:31

science experiments and physics experiments. And

8:33

my mom studied like plasma. and

8:35

like how plasma you know worked

8:37

inside of stars and stuff like

8:40

that so it was just the

8:42

wildest stuff and you would talk

8:44

to people's parent people like I talked

8:46

to my classmates or I talked

8:48

to their parents about what they're working

8:50

on is always some crazy science

8:53

thing so that was wow that was

8:55

really cool because everybody in that town

8:57

basically is they're all working on

8:59

some kind of crazy scientific

9:01

thing and so you kind of I mean I

9:03

feel like I grew up like I grew up Feeling

9:06

like you know anything was possible in

9:08

that way like yeah because the rest

9:10

of us in other communities shitty communities

9:12

or whatever we're just making that volcano

9:15

or whatever you know what I'm saying

9:17

we're doing like grassroots level bullshit you

9:19

know dang that's got to be wild

9:21

so every you see somebody just sneak

9:23

a Mahanna alley and buying a bit

9:26

of uranium and shit like that in

9:28

your neighborhood that's got to be kind

9:30

of unique. I remember there's someone from

9:32

our town who did a science fair

9:34

project called um It's called like great

9:37

balls of plasma and for the science,

9:39

literally for the science fair, this was

9:41

in like high school, for the

9:43

science fair, they made like huge

9:45

balls of plasma in their garage. And

9:47

I was like, what the heck? Like

9:50

this is, we're all just doing this

9:52

in high school? Damn. So do you

9:54

feel competitive or do you just feel

9:56

like hyper capable? Like did you feel

9:58

like kind of advanced? just in your

10:01

studies like when you were young

10:03

like you were in class like are

10:05

some of this stuffs kind of

10:07

coming easy to me? I ended up

10:09

what what I did is there

10:11

was a I ended up getting really

10:14

competitive about math in particular and

10:16

so so my sport was math which

10:18

is kind of crazy that algebra lazy

10:20

son I know I feel you

10:22

and it was because when I was

10:25

in middle school when I was

10:27

in sixth grade there was this one

10:29

math competition where if you if

10:31

you got top four in the state,

10:33

save New Mexico, then you would get

10:36

an all expense paid trip to

10:38

Disney World. And I remember as a

10:40

sixth grader, like, that was the

10:42

most motivating thing I could possibly imagine

10:44

was like an all expense paid

10:46

trip to Disney World. Yeah. And then,

10:49

did you win it? I got I

10:51

want I got fourth place so

10:53

I snuck in there, snuck in, I

10:55

and then I went to and

10:57

then we went to Florida to Disney

11:00

World and I hadn't traveled like

11:02

I didn't travel around too much growing

11:04

up like I mostly was in

11:06

New Mexico and we did some road

11:08

trips around the southwest. So I remember

11:11

getting to Florida and it was

11:13

extremely humid. It was like I never

11:15

felt what humidity felt like when

11:17

I landed in Florida. I was like

11:19

oh this feels bad. And then

11:21

I... Yeah, it definitely does. Yeah, it's

11:24

funny because I grew up inhuman, dude.

11:26

Like you would, like, you try

11:28

to shake somebody's hand, you couldn't even

11:30

land it because it was just

11:32

not enough viscos, just too much Loub

11:35

in a handshake. You'd sit there

11:37

for 10 minutes trying to land a

11:39

handshake, you know? Everybody was always dripping,

11:41

you know, you'd get really humid

11:43

over there. So then I became, so

11:46

then I became a athlete. That

11:48

was like a big part of my

11:50

identity was being a athlete and

11:52

I would. And you have to wear

11:54

a vest, like what do you guys

11:57

do? Is there an out, is

11:59

there a uniform for that or? Well

12:01

you, you need a, you have

12:03

a calculator. Okay. Everyone had their favorite.

12:05

Okay, you got that drakeau on

12:07

you, baby. I feel you keep strapped.

12:10

Yeah, you stay strapped at that thing.

12:12

And then, but outside of that,

12:14

not really. I mean, like, everyone of

12:16

their favorite, like, is pencil. You

12:18

know, you're pencil, your calculator, that was

12:21

your gear. And, but yeah, no,

12:23

I was, I was like a nationally

12:25

ranked math-like, is there... Wow, that's crazy

12:27

dude. So you go to what

12:29

other competitions you go to with that?

12:32

There's competitions, yeah. There's like, there's

12:34

like state level competitions, there's national level

12:36

competitions, there's like these summer camps,

12:38

math camp, I went to math camp

12:40

a bunch of times. uh... where you

12:43

were you convene with like-minded mathlets

12:45

okay wow fucking wizards in the park

12:47

uh... couple dudes fucking fighting over

12:49

a common denominator in the lobby that's

12:51

crazy bro but then you're but

12:53

just like any other sport like you're

12:56

like it's competitive yeah you gotta like

12:58

you gotta win and uh... and

13:00

so so you're chummy with everyone but

13:02

you're also like Like who's going

13:04

to win, you know? Yeah. No, dude,

13:07

competition's amazing, man. That's one thing,

13:09

too, that's so nice. I think about

13:11

when you're young, is like, if you're

13:14

involved in a sport or a

13:16

group or whatever it was, just that

13:18

chance to be competitive, you know?

13:20

Yeah. What were, like, some of the

13:22

competitions at the math thing, like,

13:24

what's a math competition, like, when you

13:27

get there? But once you get to

13:29

high school, it's just, you just

13:31

take a test. And you just, it's

13:33

like, the biggest one in America

13:35

is called the USMO. Bring it up.

13:38

USMO? USAMO? USAMO. OK. And it's

13:40

like, people all around the country take

13:42

this test. And there's a- United States

13:44

of American Mathematical Olympia. Yeah, there

13:46

you go. Okay, a participant must be

13:49

the US citizen or a legal

13:51

resident of the US. Okay, go on.

13:53

And then you, and then it's

13:55

a nine hour test. It's split over

13:57

two days, it's nine hours, it's nine

14:00

hours, it's four and a half

14:02

hours, and it's nine hours, and you

14:04

have six problems. So it's kind

14:06

of nerve racking, and you get in

14:08

there, you have four and a

14:10

half hours the first day, four and

14:13

a half hours the second day. Can

14:15

you cheat in between the days?

14:17

No, you because you only get three

14:19

questions the first day and then

14:21

you only get three questions the next

14:24

day and I remember the first

14:26

time I took it I was in

14:28

I think I was in eighth

14:30

grade first time I took it and

14:32

I like I was like so nervous

14:35

and I just I like brought

14:37

a thermus full of coffee and I

14:39

drank so much coffee that like

14:41

those four and a half hours felt

14:43

like like year it was like

14:45

I was so jittery the whole time

14:48

it was uh it was crazy damn

14:50

you out there rattling brother oh

14:52

integers make me rattle brother I feel

14:54

you know that dude so that's

14:56

pretty so so so you're competitive and

14:59

so you get out of there

15:01

and you're obviously I guess um admired

15:03

in the world of math probably or

15:05

that's like a thing that like

15:07

a that's like a pen that's a

15:10

feathering your cat so that helps

15:12

you get into MIT MIT right So

15:14

then, yeah, in MIT, on the

15:16

MIT admissions, they ask for your like

15:18

competitive math scores. So if you're, so,

15:21

so you knew a lot of

15:23

kids going there probably because it was

15:25

tons of kids. Yeah, yeah. Tons

15:27

of kids. It was like a reunion.

15:29

It was like math camp reunion.

15:31

Damn. Damn. Because I was wondering where

15:34

all y'all were at, dude, because we

15:36

were doing some other shit. And

15:38

so then, yeah, I was at MIT.

15:40

And then MIT is. like a

15:42

really intense school because they you know

15:45

the classes they don't fuck around

15:47

with they just really like they load

15:49

you up with tons of work and

15:51

most of the time you have

15:53

you you like they load you up

15:56

with like huge amounts of work

15:58

and they you know you you you

16:00

don't really know what's going on

16:02

initially and so you're just kind of

16:04

like just trying to make it through

16:07

but you know there's this motto

16:09

that that MIT has called ICTFP which

16:11

among the students stands for I

16:13

hate this fucking place oh yeah it's

16:15

heavy huh yeah but then but

16:17

then the school when you go they

16:20

tell you oh it stands for I've

16:22

truly found paradise so oh so

16:24

a couple differences of opinion of opinion

16:27

Yeah. Yeah. Damn. So it's so

16:29

it's really severe there. There's a lot

16:31

of work that loads you down

16:33

out of the gate. You do you

16:35

do a lot of work. But it's

16:38

it's kind of awesome. I mean,

16:40

because because I think the students really

16:42

band together like you like instead

16:44

of like I think instead of it

16:46

being competitive really, MIT is much

16:48

more about like everybody kind of like

16:51

coming together, working homework together and just

16:53

kind of like making through together.

16:55

What's that social life there? Like

16:57

are you dating? Are you going

16:59

to parties? What's that like for

17:01

you at that time? It's a

17:03

bunch of parties because people like

17:05

MIT, there's a lot of people

17:07

who like tinkering with gadgets. So

17:09

like tinkering with like, you know,

17:11

lights and big speakers and DJ

17:13

sets and all this stuff. So

17:15

actually the parties are pretty good

17:17

because they're like the production. value

17:19

is high, the production quality is

17:21

high. Damn! And what about when

17:24

the science kids come through, the

17:26

lab dogs, are they bringing, are

17:28

people making like unique drugs, was

17:30

there like designer drugs that are

17:32

being created by actual people, like

17:34

just, because, you know what I'm

17:36

saying, like my friends in college,

17:38

none of us would know how

17:40

to do that, but there may

17:42

be somebody at a... smart at

17:44

a more prestigious school that would

17:46

know how was that a thing

17:48

even or is that just there's

17:50

one there was one part of

17:52

the campus called East Campus where

17:54

it was like it was more

17:56

fringe and so there was like

17:58

at one point in the school

18:00

year they would in their courtyard

18:02

they would build a gigantic catapult

18:04

like a huge catapult like a

18:07

trebiche yeah what's the difference? I

18:09

don't know what the... Yeah, let's

18:11

get the difference here because people

18:13

need to know this anyway. People

18:15

have for decades now, people have

18:17

been. A Trebiche has like a

18:19

rope attached to the end of

18:21

it that flings it where Catapolt

18:23

just launches it. No, it was

18:25

a Catapolt then. Okay. Like a

18:27

big, because there's a big, like,

18:29

like ice cream scooper, looking thing

18:31

that would like, that would fling

18:33

it. And whether it's flinging Adderall

18:35

under the rest of the rest

18:37

of the campus. They would fling

18:39

stuff into the rest of the

18:41

Now that I'm thinking about, I

18:43

don't know what the, yeah, no,

18:45

these giant, this giant like catapult

18:47

things. Yeah. And so this was

18:49

like a event that would go

18:52

on and people would kind of

18:54

rave there? What are you saying?

18:56

Yeah, they would do this, they

18:58

would do other things. There were

19:00

like, they were into, they would

19:02

like build a lot of stuff

19:04

over there. And there would be.

19:06

Like people that ended up at

19:08

Burning Man later on. Yes, yes,

19:10

that was the burning, that was

19:12

core Burning Man, like there was

19:14

a, there was a satanic ritual

19:16

floor. Oh yeah. Yeah, like a

19:18

lot of, a lot of, like

19:20

it's fringe. Cool, it's cool. Right.

19:22

But, yeah, so there's all these

19:24

parties. We bragged at MIT that,

19:26

uh. you know, people from all

19:28

the neighboring schools, because Boston is

19:30

a huge college town, like tons

19:32

of tons. Boston's an amazing city.

19:34

Yeah, but no, MIT was fun,

19:37

but I was only there, I

19:39

was at MIT for one year.

19:41

Right, and you dropped out, is

19:43

that safe to say? Yeah. Okay,

19:45

you dropped out, and then you,

19:47

so you got into AI, into

19:49

the AI world, is that kind

19:51

of a safe bridge to say?

19:53

Yeah, yeah, yeah. Okay, and I

19:55

want to ask this, because, because

19:57

I know, because I know Mark

19:59

Zuckerberg, because, you know, forward thinking

20:01

that sort of world. Is there

20:03

something in college that you felt

20:05

like didn't nurture you or did

20:07

you just feel like this isn't

20:09

the place for me? Do you

20:11

feel like college doesn't nurture a

20:13

certain type of thinker or was

20:15

it just a personal choice? I

20:17

think for me it was like,

20:20

I was just feeling really impatient

20:22

and I don't really know why

20:24

really, but I remember like I

20:26

remember I was in school the

20:28

first year, it was really fun

20:30

and I really enjoyed it. But

20:32

then I remember, you know, this,

20:34

in the year when I was

20:36

at MIT was one of the

20:38

first, like, it was like one

20:40

of the early big moments in

20:42

AI, because it was, I don't

20:44

remember this, but there was an

20:46

AI that beat the world champion

20:48

at go. This was in 2015,

20:50

which when I was in college.

20:52

It's like a big checkerboard with

20:54

like white and black stones. It's

20:56

like, uh, and it was, uh,

20:58

yeah, this, this game, Alpha Go

21:00

versus Lisa Dole. So Alpha Go

21:02

versus Lisa Dole, also known as

21:05

the Deep Mind Challenge match was

21:07

a five game go match between

21:09

top go player Lisa Dole and

21:11

Alpha Go, a computer go program

21:13

developed by Deep Mind, played in

21:15

Seoul, South Korea, between nine. 9th

21:17

and 15th of March 2016. Alpha...

21:19

That's confusing how that's written. It

21:21

is very confusing on. You think

21:23

that... We got a... Alpha Go

21:25

won all but the fourth game.

21:27

All games were won by resignation.

21:29

The match has been compared with

21:31

the historic chess match between Deep

21:33

Blue and Gary Kasparov. Huh. The

21:35

winner of the match was stated

21:37

to win $1 million since Alpha

21:39

Go won Google Deep Mine stated

21:41

that the prize would be donated

21:43

to charities, including UNICEF and USAID.

21:45

That's just a joke. That's just...

21:47

But Lee received $150,000 for playing.

21:50

So this was a big moment

21:52

because this had never kind of

21:54

happened before? Never happened, yeah. And

21:56

it was a big moment for

21:58

AI. It was like, oh wow,

22:00

this stuff is like, it's really

22:02

happening. And so then this happened

22:04

in March, and I guess, yeah,

22:06

I dropped out, starting my company

22:08

in May. So I guess two

22:10

months after this, I was, yeah,

22:12

that's what the game looks like.

22:14

Oh, I've I've played this game

22:16

before. It's honestly. It's a really

22:18

like I'm not very good at

22:20

the game. It's a little more

22:22

fun than playing. Yeah, so unless

22:24

you're like, you know, in a

22:26

like Renaissance Fair Board games or

22:28

whatever. Yeah, yeah. Okay, so now

22:30

we got you, you're loose, dude,

22:33

you're out of the school, and

22:35

you're in the world, you see,

22:37

did that match, did realizing that

22:39

kind of like, spurn, you don't

22:41

wanna leave school? Or was that

22:43

just something that happened around the

22:45

same time? It kind of did,

22:47

basically, it was, it was, I

22:49

remember feeling like, oh wow, AI,

22:51

AI, it's happening. And this was

22:53

back in 2016, so like eight,

22:55

eight, nine years ago. Okay. And

22:57

then I felt like I had

22:59

to, you know, basically that, that

23:01

inspired me to start my company.

23:03

And I moved, I basically went

23:05

straight from, I remember, I flew

23:07

straight from Boston to San Francisco

23:09

and then started the company basically.

23:11

And that scale AI. Scale AI.

23:13

Okay. And so. Did you, had

23:15

you, had you been following AI,

23:18

like what are your kind of,

23:20

are you just like, knew like

23:22

this is where it's going, like

23:24

you just felt, there was something,

23:26

an instinct that you trusted, or

23:28

like, because that's a big thing

23:30

to do. I was stuck, so

23:32

I took all the AI classes

23:34

at MIT. Okay, so you already

23:36

learned a lot about it. Yep.

23:38

And then there was one class

23:40

where you had to, like, on

23:42

all the classes you had to

23:44

do side projects or final projects

23:46

of some kind. I wanted to

23:48

build a camera inside my refrigerator

23:50

that would tell me when my

23:52

roommates were stealing my food. Wang

23:54

boy, catching him! Wang boy! Wow!

23:56

Wow! And uh... But then, so

23:58

I worked on that and then

24:00

it was, there was one moment

24:03

where I was like, there was

24:05

a moment that clicked where I

24:07

was trying to build this thing

24:09

and then there was one step

24:11

that was like too easy. I

24:13

was like, whoa. that just worked

24:15

right there and then that happened

24:17

and then the the go match

24:19

happened and I was like this

24:21

this stuff is happening and so

24:23

I did you ever market those

24:25

refrigerates you ever actually create that

24:27

I didn't market them no I

24:29

could totally see that bro there's

24:31

a refrigerator every dorm has it

24:33

where there's a camera built in

24:35

and you just get you're in

24:37

you get a notification on your

24:39

phone you know you're like damn

24:41

add-nons got my hummus you know

24:43

but you got video of him

24:46

right there dude that's a great

24:48

idea yeah I love that was

24:50

that was that was college me

24:52

yeah okay so Alexander Wang he's

24:54

free in the world now he's

24:56

headed to San Francisco he's AI'd

24:58

up he feels the energy he's

25:00

motivated by some of the classes

25:02

he took he's motivated by some

25:04

of the classes he took he's

25:06

motivated by seeing that AI starting

25:08

to actually overtake humans, right, or

25:10

be able to compete with actual

25:12

human thinking with their chess match.

25:14

Yeah, I would, the way I

25:16

would think about it, or the

25:18

way I thought about the time

25:20

was like, this is, this is

25:22

becoming, you know, people have been

25:24

talking about AI for decades. Like,

25:26

it's kind of been always been

25:28

one of these things that people

25:31

have, have said, oh, it's gonna

25:33

happen, but it never really was

25:35

happening. And it was, you know,

25:37

it was really about artificial intelligence.

25:39

So I've always been like, um...

25:41

No, you have real intelligence, not

25:43

artificial, real intelligence. I don't, I

25:45

mean, I think it's a, it's

25:47

probably a mix, but I see

25:49

what you're saying, you know? I

25:51

do do, you know what I

25:53

thought of the other day, it

25:55

was like, what if they had

25:57

like a Mexican version, it was

25:59

like, hey, I... I don't know,

26:01

that's a good joke, but thank

26:03

you, this is a nice laugh.

26:05

This show is sponsored by Better,

26:07

better help. What are some of

26:09

your relationship green flags? That's a

26:11

good question. What are things you

26:13

notice in your relationship that are

26:16

like, yes, this is positive, this

26:18

is good, we're headed in the

26:20

right direction? We often hear about

26:22

the red flags we should avoid,

26:24

but what if we focus more

26:26

on looking for green flags in

26:28

friends and partners? If you're not

26:30

sure what they look like, therapy

26:32

can help you identify green flags,

26:34

actively practice them in your relationships,

26:36

and embody the green flag energy

26:38

yourself. I've personally benefited from therapy.

26:40

over the years. I think one

26:42

of the things that's helped me

26:44

with is just noticing some of

26:46

my patterns, noticing when I let

26:48

something small, turn into something big,

26:50

and being able to cut it

26:52

off at the past so it

26:54

doesn't happen anymore. Better help can

26:56

help you. Better help is fully

26:59

online, making therapy affordable and convenient,

27:01

serving over 5 million people worldwide.

27:03

Discover your relationships, green flags, with

27:05

better help. Visit better help.com/theo to

27:07

get 10% off your first month.

27:09

That's better help help.com/t.h.o. Are you

27:11

struggling to keep your website and

27:13

digital marketing up-to-date while running your

27:15

business? Now you can relax and

27:17

binge on unlimited web design services

27:19

with modify. Enjoy unlimited support. a

27:21

quick turnaround and your own designated

27:23

designer. Modify has done my website

27:25

for years and they're incredible. Sign

27:27

up with Modify today. They make

27:29

life easier and business better. Visit

27:31

modify.com/theo. That's M-O-D-I-P-H-Y-I-I-K-S-I-K-S-E-O for 50% off.

27:33

The last website you'll ever need.

27:35

That's modify.com/theo. What is AI? Yeah,

27:37

so so AI is all about,

27:39

you know, basically programming computers to

27:41

be able to start thinking like

27:44

humans. So, you know, traditional computer

27:46

programming, you know, it's pretty, it's

27:48

pretty bare bones. It's not very

27:50

smart. And so AI is all

27:52

about can you have, can you

27:54

build algorithms that start to be

27:56

able to. think like people and

27:58

and and replicate some of the

28:00

Like our brains are these incredible

28:02

incredible things, you know, and that's

28:04

that's evolution. That's just biology that

28:06

created our brains and so it's

28:08

all about how can we how

28:10

can we build something similar to

28:12

that or replicate it using using

28:14

computers and machines and so the

28:16

whole you know the the whole

28:18

modern AI era really started around

28:20

an area called computer vision, but

28:22

it was like how can we

28:24

first get computers to see like

28:27

humans do. So one of the

28:29

very first AI projects was this

28:31

thing called ImageNet. ImageNet and later

28:33

AlexNet. And it was basically, can

28:35

you get computer programs like given

28:37

a photo to tell you what's

28:39

in that photo? I see. So

28:41

just like a human would, like

28:43

if you showed them this. Yeah.

28:45

And you're starting to train them

28:47

to have a perspective. Yeah, train

28:49

them to, well actually, originally like,

28:51

you could, like, let's say you

28:53

took a photo of this, of

28:55

this bottle, a machine wouldn't even

28:57

be able to tell you what's

28:59

in the photo. It would just

29:01

know what the pixels were, but

29:03

it wouldn't be able to tell

29:05

you like, oh, there's a bottle

29:07

in that, in that photo. So,

29:09

you know, one of the first

29:12

AI breakthroughs was when YouTube built

29:14

an algorithm that could tell when

29:16

there were cats in their videos.

29:18

And that was like. this, you

29:20

know, in like 2012 or 2011,

29:22

this was like this mind-blowing breakthrough

29:24

that you could like, figure out,

29:26

you could like, use an algorithm

29:28

to figure out when there was

29:30

a cat inside a, inside a

29:32

video. And so, AI, it started

29:34

pretty simply with just how do

29:36

we replicate, you know, vision, like

29:38

how do you replicate like our,

29:40

basically the fact that our eyes

29:42

and our brains can process all

29:44

this imagery coming in. And that

29:46

really led to, I think one

29:48

of the first major major, use

29:50

cases of AI, which is self-driving

29:52

cars. So this was, when I

29:54

started the company in 2016. self-driving

29:57

cars were all the rage because

29:59

you know it was you know

30:01

there were all these like skunkworks

30:03

projects when you started scale AI

30:05

you mean yeah when you started

30:07

scale yeah when you when you

30:09

kind of got into AI yep

30:11

at that time self-driving cars are

30:13

the most popular things yeah yeah

30:15

that was all the rage and

30:17

so it was all about can

30:19

you know can you start building

30:21

algorithms that can drive a car

30:23

like a human would and do

30:25

it safely and do it you

30:27

know more efficiently and that way

30:29

major areas. And then now, you know,

30:31

the whole, the whole, the whole industry has

30:34

moved so fast, but then all of a

30:36

sudden we got chat GPT and we got,

30:38

you know, more advanced stuff more recently that,

30:40

that is able to talk like a human

30:43

or sort of think like a human. And

30:45

so it's really come pretty far recently, but

30:47

all of it is about how do you

30:49

build algorithms, how do you

30:51

use machines to be able to think

30:54

like a person. Okay. And is it

30:56

a pro, like say if I opened

30:58

a door, like, like, oh, we keep

31:00

the AI in there, is it a

31:02

computer, is it a program, is it

31:05

a hard drive? Like, what, like, what

31:07

is that? Yeah, yeah. So

31:09

there's two parts, there's two

31:11

parts to it. So the

31:13

first part is you need

31:15

really advanced chips. So you need

31:17

like, Like these, they're called GPUs or

31:20

sometimes called TPUs or, you know, there's

31:22

a lot of different words for it,

31:24

but you need like the most advanced

31:27

computer chips in the world. Okay. And

31:29

they, how big is each one, do

31:31

you know? Uh, like, or can you

31:34

measure it like that? They, I mean,

31:36

the biggest ones are actually like, like,

31:38

the whole chips are like, you know,

31:40

they're like a wafer kind of thing,

31:43

but then you, you, you put a

31:45

lot of them all of them all together.

31:47

these like these chips and the

31:50

biggest ones are are yeah exactly

31:52

that's a little one that's a

31:54

little one there's really big ones

31:56

okay but these are the so

31:59

this is the These are the brain

32:01

cells of it. Yep. These are the

32:03

brains behind it. Yeah. So then, yeah,

32:05

exactly. These are some, those are some

32:08

big ones. So then. So you have

32:10

to have a lot of these chips.

32:12

So you need a ton of these

32:15

chips. And that's, that's kind of the,

32:17

like, that's the physical presence. And then,

32:19

and then by the way, they take

32:21

a huge amount of energy. They're really,

32:24

because they have, you have to do

32:26

a lot of calculations. basically data centers

32:28

buildings full of like tons and tons

32:31

of those chips just in giant rows

32:33

like how big are we talking warehouses?

32:35

Yeah the biggest I mean like like

32:38

Elon's data center Colossus is like I

32:40

mean it's probably more than a million

32:42

it's definitely more than a million square

32:44

feet I mean it's like just huge

32:47

really yeah look up Colossus I've never

32:49

known this yeah yeah that yeah that

32:51

you see that building with the sunset

32:54

second row yeah there you know there

32:56

you know Or the one to the

32:58

left. Oh no. Yeah, that one. Yeah,

33:00

look, it's like a huge ass building.

33:03

What? It's huge, and all that's just

33:05

filled with chips. Have you ever been

33:07

in there? I haven't been in that

33:10

one, but I've been in some of

33:12

these, and it's just, yeah. And this

33:14

is what it looks like inside? Yeah,

33:17

basically. Yeah. So it's just rows and

33:19

rows of chips. No plants or anything.

33:21

No plants. It gets hot in there.

33:23

I bet. So the first part is

33:26

just that the that's the physical presence

33:28

and then the second part are the

33:30

algorithms so then you have like on

33:33

top of those chips you have you

33:35

have software that's running and the the

33:37

algorithms are just all are like what

33:39

you actually are telling what's the math

33:42

that you're telling to happen on the

33:44

chips and those algorithms are you know

33:46

some of the you know most complicated

33:49

algorithms that humans have ever come up

33:51

with and that's kind of the that's

33:53

kind of the software part or that's

33:56

kind of the that's the part that

33:58

like you know exists on the internet

34:00

or you can download or whatnot, and

34:02

then it has to be run on

34:05

these like huge warehouses of giant of

34:07

giant chips. Okay. So when someone goes

34:09

to like scale AI or chat GPT,

34:12

these are all AI interfaces or what

34:14

are they? Yeah, so so yeah, like

34:16

chat GPT is a is a way

34:18

to be able to talk to basically

34:21

you can talk to the algorithm. So

34:23

you can start interacting directly with the

34:25

algorithm, you can see how the algorithm

34:28

is thinking. So you could say to

34:30

this, can you describe the weather today?

34:32

Exactly. Yeah. And if you said that

34:35

to five different AI companies, or AI

34:37

companies, basically, or AI algorithms, different AI

34:39

systems, yeah. So if you said that's

34:41

five different AI systems, you might get

34:44

a little bit of a varied answer.

34:46

a little bit yeah okay you'll get

34:48

because they all are trying to have

34:51

their own style and have their own

34:53

vibe to it interesting and then what

34:55

we do what what scale AI is

34:58

all about is we've kind of built

35:00

the almost like the Uber of AI

35:02

okay so a lot of what we're

35:04

trying to do is how do we

35:07

so how do we help produce data

35:09

that is improving these algorithms and just

35:11

like how Uber there's okay you're losing

35:14

me there a little Yeah, yeah. It's

35:16

okay. But if I slow down, if

35:18

you're losing me, explain that to me

35:20

a little bit clear for me. Yeah.

35:23

So, so, okay, so with these algorithms,

35:25

one key ingredient for these algorithms is

35:27

data. So, okay, so you have the

35:30

chips and everything that are storing all

35:32

the information. Yep. They're storing the data.

35:34

And then you have the algorithms that

35:37

are helping mediate between the user and

35:39

the data. Yeah, so basically you kind

35:41

of have, yeah, you have three, you

35:43

have three key pieces. Okay. So you

35:46

have the, you have the, you have

35:48

the, you have the data, which is

35:50

like, just tons and tons of data

35:53

that, that's where the algorithms are learning

35:55

the patterns from. Okay. So these algorithms,

35:57

they aren't just like, they don't just

35:59

learn to talk randomly. They learn it

36:02

from learning to. talk from how humans

36:04

talk, right? So you need tons and

36:06

tons of data. And then you have

36:09

the algorithms which learn from all that

36:11

data, and then they run on top

36:13

of the chips. Got it. So then

36:16

one of the big challenges in the

36:18

industry is, okay, how are you going

36:20

to produce all this data? And so

36:22

this is how are you going to

36:25

get data for your? Yeah, system, like

36:27

how do you farm the best data?

36:29

How do you, exactly, how do you

36:32

build, how do you build all that

36:34

data? And how do you do that

36:36

in the most effective way? How do

36:38

you build new data? So clean data,

36:41

because what if you get a bunch

36:43

of data in there, there's just a

36:45

bunch of advertisements in bullshit, will that

36:48

affect the output? Yeah, that definitely affects

36:50

the output. So data, so data is,

36:52

you know, some people say like, like,

36:55

data is the new oil, it's how

36:57

the algorithms are learning everything they're learning.

36:59

Like anything that the algorithms know or

37:01

learn or say or do, all that

37:04

has to come from the data that

37:06

goes into it. Okay, so if I

37:08

ask the, if I ask a system,

37:11

an AI system, a question, or ask

37:13

it to help me with something, help

37:15

me to design something or to curate

37:17

an idea, it's gonna use the data

37:20

that it has within it to. respond

37:22

to me and help me and help

37:24

give me an answer that I can

37:27

use. And it's only, and the data

37:29

it has in it is only based

37:31

upon the data that is put into

37:34

it. Exactly. Yeah, yeah. Yeah. So then,

37:36

so yeah, it's kind of, so then

37:38

we don't, you know, we don't spend

37:40

enough time talking about where it is,

37:43

you know, how are you going to

37:45

get this data? How are you going

37:47

to keep making new data? So the

37:50

angle that we took at scale was

37:52

to kind of, turn this into an

37:54

opportunity for people. So we're, you know,

37:56

we're kind of like the Uber for

37:59

AI. So just like how Uber you

38:01

have, you know, riders and drivers, for

38:03

us, we have, you know, we have

38:06

the AI. systems, you know, the algorithms

38:08

that need data, and then we have

38:10

a community of people, a network of

38:13

people who help produce the data that

38:15

go into the system. They get paid

38:17

to do that. Oh, so they're almost

38:19

data farming, like creating good data? Creating

38:22

good data, exactly. And it's huge. It's

38:24

like, so we do this through our

38:26

platform called Outlier. people, contributors, we call

38:29

them, contributors on Outlier, earned about 500

38:31

million dollars total across everybody in the

38:33

U.S. that's across 9,000 different towns. And

38:36

so it created a lot of jobs,

38:38

a lot of jobs, okay. And so

38:40

what would, okay, so scale was your

38:42

company, it's an AI system. Yep. Is

38:45

that right? So we, yeah, I mean,

38:47

yeah, Scaly is an AI system. Yep.

38:49

And then Outlier. is a separate company

38:52

that works with it. Yep. And that

38:54

is where you are hiring people. We,

38:56

yeah, we, we basically. To pull in

38:58

data. Yeah, we, we build this platform

39:01

that anybody, you know, a lot of

39:03

people, frankly, all around the world, but

39:05

Americans too, can log on and, and

39:08

help build data that goes into the

39:10

algorithms and get paid to do so.

39:12

So how does a user do that?

39:15

Like what is an example of somebody

39:17

who's helping build data for an AI

39:19

database? Yeah, let's say you're a nurse.

39:21

Like you're a nurse with like tons

39:24

of tons of experience. So you know

39:26

a lot about how to take care

39:28

of people and take care of people

39:31

who are sick or have issues and

39:33

whatnot. And so you could log on

39:35

to the system and our platform and

39:37

you could see that the algorithm. is,

39:40

you know, let's say you ask the

39:42

algorithm like, hey, I have a, you

39:44

know, I have a pain in my,

39:47

in my stomach, what should I do?

39:49

And you notice that the algorithms. says

39:51

the wrong thing. Like the algorithm says,

39:54

oh, just, you know, hang out and,

39:56

you know, it'll go away. And you

39:58

know, as a nurse, like, that's wrong.

40:00

Like, I, you know, you have to,

40:03

you have to go to the emergency

40:05

room because you might have a appendicitis

40:07

or you might have, you know, you

40:10

might have something really bad. And so

40:12

you would, as a nurse, you would

40:14

go in and you'd basically, this continual

40:16

process of and there's versions of that

40:19

for whatever your expertise is or whatever

40:21

you know you know more about then

40:23

anything everything everything exactly so and so

40:26

people get paid for that yeah they

40:28

get paid and how do you know

40:30

if their information is valuable or not

40:33

well we so we don't want spam

40:35

obviously so we we we have a

40:37

lot of systems to make sure that

40:39

people aren't spamming and that like you're

40:42

saying it's not it's not you know

40:44

it's not garbage in that's going into

40:46

the into the algorithms so we have

40:49

you know we have kind of like

40:51

people check the work of other people

40:53

to make sure that the AI systems

40:55

are really good and we have some

40:58

like automated systems that that check this

41:00

stuff but uh but for the most

41:02

part it's it's like it's really broad

41:05

like we want experts in anything everything

41:07

shellfish train tracks whatever yeah everything. Childhood,

41:09

death or whatever. Yeah, totally. Stars, galaxies,

41:12

whatever. Yeah, animals. Yeah. Yeah. So, wow.

41:14

So it's kind of like your data

41:16

is almost like an ocean or a

41:18

body of water and you, different places

41:21

are going to be able to keep

41:23

their body of water cleaner or dirtier

41:25

and different infections could get in, different

41:28

spyware, all types of stuff. So, and

41:30

if you have really a clean body

41:32

of water, then you're going to be

41:34

able to offer a clean data or

41:37

a certain type of data to people

41:39

who are using your AI platform. Exactly.

41:41

Does that make sense or not? Yeah,

41:44

yeah, totally. And our job, like, how

41:46

do we make sure that this body

41:48

of water is as clean as possible

41:51

and... we fill it up as much

41:53

as possible that it has as much

41:55

information about everything across the globe. Wow,

41:57

so is there almost a race for

42:00

information right now in a weird way

42:02

or no? Is that not it? There

42:04

a little bit. Yeah, I think that

42:07

there's a well, there's a race for.

42:09

Like how are different AI systems competing

42:11

against each other? And sorry to interrupt

42:13

you. Yeah, no, no. So there's, it

42:16

goes back to the three things I

42:18

mentioned. So there's kind of like three

42:20

dimensions that they're all competing as one

42:23

other. chips. So who can, who is

42:25

the most advanced chips, who is the

42:27

biggest buildings of chips, like who is

42:30

the most chips that they're utilizing, data,

42:32

so the kind of body of water,

42:34

whose body of water is better, cleaner,

42:36

you know, healthiest, biggest, etc. And then

42:39

the last is algorithms. So who's, and

42:41

this is where the scientists really come

42:43

in. It's like, okay, who's coming up

42:46

with the cleverest algorithms or who has

42:48

like a trick on an algorithm that

42:50

somebody else doesn't have, like who's doing

42:53

that to basically make the AI learn

42:55

better off of the data that it

42:57

has. Wow, God, I'm in the future

42:59

right now. That's so wild. Man, it's

43:02

just, it's so crazy. And I think

43:04

AI scares people because the future scares

43:06

people, right? It's like, that's one of

43:09

the scariest things sometimes is the future.

43:11

So I think a lot of times

43:13

you associate, because a lot of some

43:15

people mention AI, there's a little bit

43:18

of fear it seems like from people,

43:20

it seems like, from people. Yeah. There's

43:22

fear that it's gonna take jobs, there's

43:25

fear that it's gonna take over our

43:27

ability to think for ourselves, right? And

43:29

not a lack of knowledge because you

43:32

didn't want to know just because you

43:34

don't know. Or that you're dumb, but

43:36

just because you don't know. What are

43:38

positive things that we're going to see

43:41

with AI, right? I want to start

43:43

there. Yeah. So I think first, like,

43:45

we don't, the AI industry, we don't

43:48

do the best job explaining this. And

43:50

I think some. Sometimes we make it

43:52

seem all sci-fi and genuinely we're part

43:54

of the problem in making it seem

43:57

scary, right? But one thing, for example,

43:59

is like, I think AI is actually

44:01

gonna create a ton of jobs and

44:04

that story is not told enough, but

44:06

you know. these jobs that we're producing

44:08

are this this sort of this opportunity

44:11

that we're providing on our platform outlier

44:13

like that's only gonna grow as AI

44:15

grows so because you have to have

44:17

new data you have to have new

44:20

data and the only place you can

44:22

get new data is from people will

44:24

at a certain point would the system

44:27

be able to create it can probably

44:29

matriculate data or matriculate is that with

44:31

birthing or what is that yeah it

44:33

just moves through like Okay, it can

44:36

probably like quantify or in and give

44:38

you answers but it can AI create

44:40

new data? No, so I think well

44:43

it can do a little bit of

44:45

that so it can AI can help

44:47

itself create its own data and and

44:50

help itself a little bit but ultimately

44:52

most of the progress is going to

44:54

come from you know people were able

44:56

to help. really the model get better

44:59

and smarter and more capable at all

45:01

these all these different areas. Yeah I

45:03

didn't see that part of it I

45:06

didn't understand that we are the ones

45:08

who are giving it information and since

45:10

we're going to continue to learn I

45:12

would assume that we would be able

45:15

to help it to help it learn.

45:17

Yeah and the world's going to keep

45:19

changing and we're going to need to

45:22

be able to keep teaching the algorithms

45:24

keep teaching the world's about how the

45:26

world's changing. So you know this is

45:29

actually a a big thing that I

45:31

think most people don't understand. The people

45:33

who are getting the opportunity now are

45:35

earning money from it, see it. But

45:38

as AI grows, there's actually going to

45:40

be tons of jobs created along the

45:42

way and tons of opportunity for people

45:45

to help improve AI systems or control

45:47

AI systems or overall sort of be

45:49

a part of the technology, not just

45:51

sort of disenfranchised by it. Okay, so

45:54

like yeah, so what were what are

45:56

you, do you feel like are other

45:58

ways like if you had to look

46:01

into the future a little bit, right?

46:03

So you have the fact that people

46:05

are going to be able to add

46:08

more data, right? And add nuances to

46:10

data, right? And probably humanize data a

46:12

little bit. Yeah, totally. And then you're

46:14

also going to have what I think

46:17

you're going to have a lot of,

46:19

a lot of jobs around, you know,

46:21

doing all these little things throughout the

46:24

world, who's gonna keep watch of those

46:26

AI? And who's gonna make sure that

46:28

those AI aren't doing something that we

46:30

don't want them to do? So almost

46:33

like managing the AIs and keeping watch

46:35

over all the AI systems, that's gonna

46:37

be another thing that we're gonna have

46:40

to do. And then. And then it's

46:42

just somebody to kind of guide the

46:44

river a little bit. Yeah, at certain

46:47

point, guide the stream, stay in there,

46:49

watch, make sure that answers are correct,

46:51

make sure the information is honest. Yeah,

46:53

yeah, like, like I think, for example,

46:56

you know, we're not gonna just have

46:58

AIs going around and, you know, you

47:00

know, buying stuff and doing crazy things

47:03

and like, you know, we're gonna, we're

47:05

gonna keep it controlled, right? Like as

47:07

a society, I think we're gonna keep

47:10

it controlled as a technology. And I

47:12

think there's gonna be a lot of

47:14

jobs for people to make sure that

47:16

the AI doesn't go out and do

47:19

crazy things that we don't want it

47:21

to do. Right so we want to

47:23

be so you're gonna need managers you're

47:26

gonna need facilitators. Yeah exactly what are

47:28

things that AI will alleviate like what

47:30

are things that like will it eventually

47:32

be able to have enough information like

47:35

our data where where it can like

47:37

cure diseases and stuff like that like

47:39

is that a realistic thing? Yeah that's

47:42

super real like cancer even yeah cancer

47:44

yeah heart disease like all these all

47:46

these diseases Yeah. Hands are on his

47:49

heels. But, but. around us here I

47:51

think there's still some smoke in the

47:53

air no but seriously I think that

47:55

AI one thing that we've seen which

47:58

is this is kind of wild but

48:00

AI understands like molecules and biology better

48:02

than humans do actually because it's like

48:05

like there's this there's this thing in

48:07

AI where you know it used to

48:09

take a like a PhD biologist like

48:11

you know five years to do something

48:14

that the AI can can just do

48:16

in you know a few minutes right

48:18

and that's because the like just the

48:21

way that molecules and biology and all

48:23

that works is something that that AI

48:25

happens to be really good at and

48:28

that's gonna help us ultimately cure diseases

48:30

find you know pharmaceuticals or other treatments

48:32

for these diseases and ultimately help humans

48:34

live longer. Because that's very data driven,

48:37

right? Like it's very specific, it's very

48:39

mathematic. Exactly. Yeah, yeah. So it's going

48:41

to be a huge tool for us

48:44

to cure disease, for us to help

48:46

educate people, for us to, you know,

48:48

there's a lot of really exciting uses

48:50

for AI. But I think the kind

48:53

of, I think the thing that, um,

48:55

will touch most people in their lives

48:57

is it's really gonna be a like

49:00

a tool that will help you You

49:02

know make all of your sort of

49:04

Make all your dreams kind of become

49:07

reality if that makes sense. So so

49:09

I think one of the things that

49:11

AI is gonna be really awesome for

49:13

is like you know today if I

49:16

I have like a million ideas, right?

49:18

I have like you know thousands and

49:20

thousands of ideas and I only have

49:23

so much time so I can only

49:25

really do you know a few of

49:27

them at a time. And most of

49:29

the ideas just go to die, right?

49:32

Yeah, they do, huh? Yeah, it's a

49:34

bummer. It's a huge bummer. Yeah. And

49:36

I think a lot of people, you

49:39

know, for whatever reason, they may have

49:41

some of the best ideas ever, but

49:43

they just, you know, they're too busy

49:46

or they have like other shit going

49:48

on in their lives, they can't make

49:50

those ideas happen. And it's great with

49:52

sometimes when people are able to make

49:55

the leaps and make them happen and

49:57

devote themselves to their dreams, but that

49:59

doesn't happen enough today. And one of

50:02

the things that AI is going to

50:04

help us do, is it's going to

50:06

help us turn these ideas into reality

50:08

much more easily. So, you know, you

50:11

can, like, you know, you're making a

50:13

movie. Let's say you have another movie

50:15

idea. ultimately I think you'll be able

50:18

to tell an AI hey I have

50:20

this idea for a movie what could

50:22

that look like you know maybe draft

50:25

up a script also who are the

50:27

people who can help you know fund

50:29

this idea like who those people be

50:31

can help reach out to them and

50:34

then like you know who should we

50:36

cast in it basically help make the

50:38

whole thing a you know instead of

50:41

those like daunting thing that these big

50:43

projects are usually so daunting you really

50:45

don't know where to get started you

50:47

kind of need a person to help

50:50

you get through them instead of that

50:52

AI will help you get through it

50:54

and like help do a lot of

50:57

the the sort of less glamorous work

50:59

to make them a reality. Wow. So

51:01

I could say for example like like

51:04

AI I would like to shoot maybe

51:06

I'm thinking about creating an idea or

51:08

shooting a film in this area or

51:10

it's like this it's gonna take place

51:13

in this type of a place I

51:15

could give it like a Outline of

51:17

the characters like what they look like

51:20

their ages and some description could you

51:22

help give me? Possible potential actors or

51:24

something within a certain price range that

51:27

I can maybe cast for that Yeah,

51:29

could you help give me like locations

51:31

around the country that would fit that

51:33

backdrop? Yep. Could you? List me all

51:36

the talent agencies that I could reach

51:38

out to and you can kind of

51:40

just put those things in and then

51:43

you would have Sort of a a

51:45

a bit of a guidebook at that

51:47

point that would make your yeah, what

51:49

before is something that felt extremely daunting

51:52

Yeah, shit in two minutes, you know,

51:54

and then you put it in the

51:56

AI gives you information back in a

51:59

few minutes and maybe you're... And I

52:01

think I think over time it'll also

52:03

be able to start doing a lot

52:06

of the legwork for you. So it'll

52:08

be able to reach out to people

52:10

for you. It'll be able to you

52:12

know figure out the logistics. It'll be

52:15

able to to book things for you.

52:17

Like it'll be able to basically help

52:19

do all the legwork to make it

52:21

make you know whatever it is into

52:23

a reality. So very much an assistant

52:25

in the eye world as agents but

52:27

you know it's just... It'll be something

52:30

that will help you, you know, humans

52:32

are going to be in control.

52:34

Humans are ultimately going to be

52:36

telling the AIs what they want it to do.

52:38

And then, you know, hey, what do we

52:40

want these AIs do? It's ultimately going

52:42

to be to like help us execute

52:44

on and accomplish all these ideas that

52:46

we have. So a genius could be three

52:49

or four X. If somebody's like a

52:51

genius in something in some realm or

52:53

space of thought, you could three or

52:55

four X them because they like multiply

52:57

their output. from their own brain because

52:59

they could have something really helping them

53:01

get done a lot of the like

53:03

the early work on things and maybe

53:05

some of the most severe work. Yeah totally.

53:07

Like I think one thing that I always

53:09

feel like kind of sucks is that if you

53:11

have a director you really like they're only

53:14

going to make a movie once every couple

53:16

of years. So even if you have a

53:18

director that you like think is amazing. You

53:20

know they just it's hard for them

53:22

to make that many movies like because

53:25

it just takes so much time and

53:27

effort and you know there's somebody bottlenecks

53:29

and stuff so in a future with you

53:31

know more advanced AI systems they could they

53:33

could just churn them out. and they can

53:36

make so many of their ideas into reality.

53:38

And I think that's true not only in

53:40

creative areas, but it's kind of true across

53:42

the board. Like, you know, you can start

53:44

new businesses more easily. You can, you know,

53:46

you can make various creative projects that you

53:49

have happened more easily. You can make like,

53:51

you can, you can finally plan that event

53:53

that you and your friends have been talking

53:55

about for, you know, years and years. So,

53:57

it can just like, really help you, you

53:59

know. we think about was like giving humans

54:02

more agency giving humans more sort of

54:04

sovereignty and just and just enabling humans

54:06

to get way more done. Yeah that's

54:08

a great way I like some of

54:10

this thought because yeah I could be

54:13

like my fantasy football group and I

54:15

we do a we do a draft

54:17

every year in a different location you

54:19

know and shout out PA that's our

54:22

fantasy league Jayrod everybody that's in it

54:24

for the past 17 years we've flown

54:26

to a city each year and done

54:28

a draft in person right. But I

54:30

could say hey We have 10 guys

54:33

we want to go to a nice

54:35

place. We want there to be a

54:37

beach This is the age group of

54:39

our group You know these would kind

54:42

of be the nights and you know

54:44

just to really give me like a

54:46

Just a nice plan of hey, here's

54:48

ten possibilities right something like that and

54:50

then even more like with a movie

54:53

This is what I were you'd run

54:55

into say if I'd be like okay.

54:57

I have two main characters and this

54:59

is kind of what I would like

55:01

to happen. Could you help me with

55:04

a first act? of a three-act movie.

55:06

How do you know everybody just doesn't

55:08

get the same movie then? Like that's

55:10

what I would start to worry that

55:13

everything that you have at Creative is

55:15

all gonna be very similar. Yeah, so

55:17

then I think this is where it

55:19

comes into like, this is where human

55:21

creativity is gonna matter, because it's gonna

55:24

be about, then it's about like, okay,

55:26

what is the, how am I, you

55:28

know, How am I directing the AI

55:30

system? Like what are my, what are

55:33

the tricks I have to make the

55:35

AI give me something that's different and

55:37

unique? And that's not different at all

55:39

from, you know, how creative stuff works

55:41

today, like even on social media or

55:44

anywhere, you know this. Like, yeah, you

55:46

still had your own spice to it.

55:48

You need, yeah, you need an angle,

55:50

you always need to like have something

55:52

that's gonna make it like different and

55:55

interesting and new. So that's not gonna

55:57

change. Like, humans are always gonna have

55:59

to figure out to figure out to

56:01

figure out. based on where culture is

56:04

at, based on where, you know, what

56:06

the dialogue is, what the, what the

56:08

discourse is, all that kind of stuff.

56:10

What is my fresh take? That's where,

56:12

that's really one of the key things

56:15

that humans are going to have to

56:17

keep doing no matter what. Right. And

56:19

so. Some of the, like a lot

56:21

of films and books, a lot of

56:24

it is, there's just like a problem,

56:26

there's maybe a information you learn, there's

56:28

a red herring, and then there's a

56:30

solution. Like, that's a lot of stories,

56:32

right? So, if something just gave you

56:35

the basis and then you go through

56:37

and make everything your own, because a

56:39

lot of things, we don't, there's only

56:41

so many templates for things. So say

56:43

for example, say you might need to

56:46

hire people at a company then that

56:48

would help direct your AI, like somebody

56:50

who's good at managing AI, and giving

56:52

it the best prompts or the best

56:55

way to ask it questions, to get

56:57

the perfect feedback for your company. So

56:59

those would be new jobs, those would

57:01

be actual new people you would need.

57:03

Yeah, so tons of new drops, well

57:06

first I think just like helping to.

57:08

you know, kind of what we're talking

57:10

about before. These jobs around helping to

57:12

improve the data and contribute to the

57:15

AI systems, that's just going to keep

57:17

growing for a long, long time. And

57:19

then as AI gets better and it

57:21

gets used in more areas, then there

57:23

are going to be a lot of

57:26

jobs that pop up just exactly as

57:28

you're saying, which is how do you,

57:30

what's the best way to use this

57:32

as a tool, what's the best way

57:34

to leverage it to, you know, actually

57:37

make some of these applications or you

57:39

know, whatever you want to build a

57:41

reality. And then there's going to be

57:43

folks who need to, once those are

57:46

built, like, how do you keep improving

57:48

it? How do you keep making it

57:50

better? How do you keep making it

57:52

better? How do you keep, how do

57:54

you keep it fresh? How do you

57:57

keep it, keep it going? And then

57:59

how do you also make sure it

58:01

doesn't do anything bad, right? Like how

58:03

do you make sure that the AI

58:06

doesn't accidentally spam a million people, a

58:08

million people or whatever it's. like operating

58:10

in a good way. Fahim and more

58:12

I was watching him bring up a

58:14

picture of Fahim. This is one of

58:17

the funny this guy this is the

58:19

most cre comedian in America. Undeniable. He

58:21

is so funny. He had, and everybody

58:23

would say, he's, he's one of the

58:26

few comedians that everybody goes in there

58:28

to watch him perform. He had a

58:30

bit the other night, he talks about

58:32

he got into a Waymo, right? Yeah.

58:34

A car, and I see so many

58:37

Waymo's now, which are cars that are

58:39

just, nobody's in them, you know, but

58:41

they're going, right? He had this bid,

58:43

he's like he got in Owemo and

58:45

it started complaining about its life to

58:48

him. He's like, I've had a shitty

58:50

week, like the car's just talking to

58:52

him and he's like, now I, no

58:54

matter what, I still have to talk

58:57

to my shitty driver. If you get

58:59

a chance to see him though, that

59:01

guy is, he's fascinating. But like, so

59:03

what companies right now should kind of

59:05

look to hire an AI guy? Because

59:08

we've been thinking about it like, like,

59:10

We had Cat Williams on a last

59:12

week and we're like, hey, can you

59:14

create visuals of Sugnight and Cat Williams

59:17

riding bicycles down Sunset Boulevard, right? And

59:19

this is the one they sent back

59:21

a little while later. There you go.

59:23

Christopher Follino is the guy's name, F-O-I.

59:25

And this is just, this is right

59:28

where I say, by the comedy store

59:30

on Sunset Boulevard. There you go. I

59:32

mean, this looks like it's out of

59:34

a movie kind of. Yeah, totally. I

59:36

mean, and the guy did that in

59:39

a little, in just a little bit

59:41

of time. What types of people would

59:43

you get to, I mean, this is,

59:45

that's you. Yeah. If I was healthier

59:48

if I had healthier gums too. But

59:50

what about this? What kind of, like

59:52

what companies right now, what job spaces,

59:54

because we want to get an AI

59:56

guy, right? But I don't really, my

59:59

brain is like, well, what do they

1:00:01

do? How do I, how would, I

1:00:03

keep them, I keep them busy? you

1:00:05

know I could get them to make

1:00:08

some animations and ideas but what type

1:00:10

of people need an AI person right

1:00:12

now do you feel like? I kind

1:00:14

of think about AI it's kind of

1:00:16

like the internet where you know eventually

1:00:19

everybody's going to need to figure out

1:00:21

how to how to utilize it how

1:00:23

to best how to best use it

1:00:25

for their industry or whatever they do

1:00:27

how to make it them more efficient

1:00:30

like it's something that I think everybody

1:00:32

is going to is going to need

1:00:34

to adopt at some point so you

1:00:36

know might as well start earlier because

1:00:39

eventually just like how you know every

1:00:41

Basically, every company has to figure out

1:00:43

how to use the internet well and

1:00:45

how to be smart about, you know,

1:00:47

the internet and digital stuff. Every company

1:00:50

is going to have to be smart

1:00:52

about AI, how to use AI, how

1:00:54

to make, how to have a unique

1:00:56

twist on it so that, you know,

1:00:59

their stuff stands out relative to other

1:01:01

people's. That's going to be really important.

1:01:03

But so we see, I mean, in

1:01:05

our work, you know, we work with

1:01:07

all these. all these big companies in

1:01:10

America and we see it everywhere from

1:01:12

you know we worked at Time magazine

1:01:14

on some stuff and then we worked

1:01:16

with Toyota on some stuff for their

1:01:19

cars and we worked with you know

1:01:21

large pharmaceutical companies for the biology stuff

1:01:23

we were talking about large hospitals for

1:01:25

you know helping to treat patients like

1:01:27

really it's across the board and I

1:01:30

think that goes for you know, obviously

1:01:32

these like really big businesses, but also

1:01:34

for smaller businesses, you know, there's always

1:01:36

interesting ways to utilize it to to

1:01:38

provide a better product or a better

1:01:41

experience or better content for, you know,

1:01:43

whoever you need to do that for.

1:01:45

Yeah, because I guess right now we're

1:01:47

like there's certain moments like hey, let's

1:01:50

animate this moment or see what AI

1:01:52

would look like it adds some visual

1:01:54

effects to some of our episodes So

1:01:56

that's something we like to do to

1:01:58

just be fun and creative I would

1:02:01

like to maybe create like some sort

1:02:03

of an animated character We already have

1:02:05

a great animator and we want to

1:02:07

keep that but to have an AI

1:02:10

space where it's like, you know, because

1:02:12

they have something they had a little

1:02:14

cat the other day or something and

1:02:16

he was going to war And I

1:02:18

was like damn dude, this is it

1:02:21

was AI, you know, totally and they

1:02:23

had a baby was getting in a

1:02:25

taxi And I was like this shit

1:02:27

is illegal, you know, I don't know

1:02:29

if it's illegal or not, but it

1:02:32

seems You know, it doesn't seem it,

1:02:34

you know, it's definitely a tool for

1:02:36

storytellers, right? Like it's right, it's it'll

1:02:38

help people with a creative vision or

1:02:41

or one war cats. Yeah, that's a

1:02:43

story right there special forces cats Would

1:02:45

YouTube recognize this? Damn, brother, if they

1:02:47

show up, dude. Wow. This honestly, it

1:02:49

looks, it looks bad ass. And it

1:02:52

looks like cats have been waiting to

1:02:54

do this a long time. Yeah. That's

1:02:56

the crazy shit about cats. You look

1:02:58

in their eyes. Now their outfits finally

1:03:01

match the look in their eyes. That's

1:03:03

what it feels like, dude. That's, uh,

1:03:05

that's, Fremen, Fremen cats. Oh my gosh,

1:03:07

yeah. So that's gonna get alarming, dude.

1:03:09

That's gonna get hella alarming. That's a

1:03:12

movie right there. It is, but it's

1:03:14

almost like you could make, like, that's

1:03:16

what I wanna get. I wanna get

1:03:18

somebody to help us think, hey, don't

1:03:20

make these little segments that we can

1:03:23

take our creativity. And instead of me

1:03:25

thinking, man, I gotta write this huge

1:03:27

crazy script for just kind of small

1:03:29

things, you know, little moments. Yeah. Can

1:03:32

you help me make this happen make

1:03:34

this happen? AI will just be something

1:03:36

that we turn to to help make

1:03:38

our dreams happen like we help make

1:03:40

our ideas and our dreams and our

1:03:43

you know whatever we want to do

1:03:45

happen more easily got it that's like

1:03:47

at the core what it'll be yeah

1:03:49

who's the who's the current leader in

1:03:52

AI development like is it America is

1:03:54

it China is it is real is

1:03:56

it trying to think of another super

1:03:58

power Russia maybe or yeah who is

1:04:00

it Taiwan I know has a lot

1:04:03

of the chips yeah and they manufacture

1:04:05

a lot of the chips over there

1:04:07

and does it matter what country leads

1:04:09

in AI or does it just matter

1:04:11

the company like scale or another AI

1:04:14

company so yeah it's so today America

1:04:16

is in the lead but But China

1:04:18

as a country is sort of hot

1:04:20

on our tails. Like there was all

1:04:23

that news about Deep Seek a couple

1:04:25

weeks ago. And Deep Seek still in

1:04:27

most places around the world is the

1:04:29

number one most downloaded app. You know,

1:04:31

it's downloaded a ton. you know everywhere

1:04:34

around the world frankly and is a

1:04:36

Chinese AI system right and so it's

1:04:38

starting to rival a lot of the

1:04:40

American AI systems also because it's free

1:04:43

and you know it uh it kind

1:04:45

of like shocked the world so right

1:04:47

now if you kind of look at

1:04:49

it US and China are are a

1:04:51

little bit neck and neck maybe the

1:04:54

US and America's like a little bit

1:04:56

ahead and you kind of like look

1:04:58

at If you go back to each

1:05:00

of the three pieces that I talked

1:05:03

about, so the chips and the computational

1:05:05

power, the data, and the algorithms, if

1:05:07

you were to rack and stack US

1:05:09

versus China on each one of those,

1:05:11

we're probably, we're ahead on the computational

1:05:14

power because the United States is the

1:05:16

leader at developing the chips and most

1:05:18

of the most advanced chips are American

1:05:20

chips. They probably beat us out on

1:05:22

data. because China, they've been investing into

1:05:25

data for a very long time as

1:05:27

a country. And then on algorithms, we're

1:05:29

basically neck and neck. So it's a

1:05:31

pretty tight race. And to your question

1:05:34

about, does it matter? Or what does

1:05:36

this mean? I think it's actually going

1:05:38

to be one of the most important,

1:05:40

you know. questions or most important races

1:05:42

over time is, is it US or

1:05:45

Chinese AI that wins? Because, you know,

1:05:47

AI is more than just being a

1:05:49

tool that, that, you know, we can

1:05:51

all use to make our, you know,

1:05:54

build whatever we want to or make

1:05:56

whatever ideas we want to happen. It's

1:05:58

also, you know, it's a, it's a

1:06:00

cultural staple, right, you know, if you

1:06:02

talk to an AI, that AI is

1:06:05

kind of a reflection of. our culture

1:06:07

and our values and all that stuff.

1:06:09

So in America, we value free speech

1:06:11

and, you know, the AIs are, you

1:06:13

know, are built to support that. Whereas

1:06:16

in, in China, there's, there isn't free

1:06:18

speech. And so, you know, if, if

1:06:20

the Chinese AIs are the ones that

1:06:22

take over the world, then all these

1:06:25

Chinese ideologies are going to become exported

1:06:27

all around the world. And so first

1:06:29

is there's a couple dimensions here that

1:06:31

I think matter. So first is just

1:06:33

the cultural element, which is like, do

1:06:36

we want kind of democracy and free

1:06:38

speech to be the cultural AI that

1:06:40

wins? Or do we want sort of

1:06:42

the more, you know, frankly, totalitarian AIs

1:06:45

in China to be the ones that

1:06:47

win? And then there's sort of the

1:06:49

There's like the, you know, you start

1:06:51

getting to economically. So AI is going

1:06:53

to be something that helps all the

1:06:56

companies in the United States thrive. And

1:06:58

so if the USAI wins, then we're

1:07:00

going to, you know, the economy will

1:07:02

grow faster. We're going to have more

1:07:04

and more opportunity. You know, the country

1:07:07

will still be better and better and

1:07:09

better and better. And the economy will

1:07:11

keep growing. We're versus if Chinese AI

1:07:13

wins, then Chinese economy is going to

1:07:16

grow way faster than the American economy.

1:07:18

the economic piece. And then lastly, there's

1:07:20

the, there's kind of the warfare piece,

1:07:22

right? And, you know, AI, we haven't

1:07:24

really talked about it, but has clear

1:07:27

potential to be used as a military

1:07:29

technology. And we don't want, you know,

1:07:31

we don't want another country to have,

1:07:33

because they have better AI to have

1:07:36

a much stronger military than, and then

1:07:38

America's. So, like, how would they do

1:07:40

that? How would they have a better

1:07:42

AI or how they use it to?

1:07:44

have a better military. Yeah, how would

1:07:47

they use it to have a better

1:07:49

military? Like why is that kind of

1:07:51

a concern or potential concern? Yeah, so

1:07:53

one of the things that's been happening

1:07:55

over the past, you know, decade for

1:07:58

sure is lots of hacking, cyber hacking

1:08:00

going on. So, you know, in America,

1:08:02

even recently, we had this huge cyber

1:08:04

hack called salt typhoon, where the Chinese

1:08:07

hacked are... telecommunications companies so hacked for

1:08:09

the phone companies damn they did yeah

1:08:11

and they got it and they got

1:08:13

all sorts of crazy data as a

1:08:15

result of that. Oh, they know I'm

1:08:18

a pervert, I'll tell you that. Yeah,

1:08:20

look at this, this happened in 2020.

1:08:22

Salt typhoon is widely understood to be

1:08:24

operated by China's ministry, state of security.

1:08:27

It's foreign intelligence service and secret police.

1:08:29

Chinese embassy denied all allegations saying it

1:08:31

was unfattening irresponsible smears and slanders. High

1:08:33

profile cyber espionage. In 2024, U.S. officials

1:08:35

announced that hackers affiliated with Salt typhoon

1:08:38

had accessed the computer systems of nine

1:08:40

U.S. telecommunications companies later acknowledged to include

1:08:42

Verizon, AT&T, T-Mobile, Spectrum, Lumen, Consolidated Communications,

1:08:44

and Winstream. The hackers were able to

1:08:47

access metadata of users' calls and text

1:08:49

messages. Fuck! Oh, we're fucked, I am.

1:08:51

You seem good. Including date and time

1:08:53

stamps, source and destination IP addresses. Ah

1:08:55

shit, keep going, keep going. And phone

1:08:58

numbers, some of our million users, most

1:09:00

of which were located in Washington DC,

1:09:02

good. Light them up, dude. In some

1:09:04

cases, the hackers are able to obtain

1:09:06

audio recordings, a telephone calls made by

1:09:09

high profile individuals. Such individuals reportedly included

1:09:11

staff of the Kamalaris 2024 presidential campaign

1:09:13

as well as phones belonging to Donald

1:09:15

Trump and JD Vance According to Deputy

1:09:18

National Security Advisor and Newberger a large

1:09:20

number of the individuals whose data was

1:09:22

directly accessed were government targets of interest.

1:09:24

Wow. Yeah, that's free. So do you

1:09:26

think this also that that whole thing

1:09:29

could be not real and it's just

1:09:31

a story that was created? That

1:09:33

seems pretty real, because there's real,

1:09:35

like, I mean, there's like 20

1:09:37

stories where the Chinese have hacked

1:09:39

American systems. Like they hacked, this

1:09:41

was, this must have been close

1:09:43

to 10 years ago now, but

1:09:46

the Chinese hacked the database in

1:09:48

America that stored all of the

1:09:50

clearances. So they hacked in, they

1:09:52

managed to hack into knowing who

1:09:54

are literally. all of the Americans

1:09:56

who have security clearance. Oh, security

1:09:58

clearance. Yeah, so I thought you

1:10:00

know what was on sale. I

1:10:02

was like, who cares? That's great

1:10:04

though. Damn, oh damn. So they

1:10:06

knew everybody who had, who knew,

1:10:08

who knew, who knew, who knew

1:10:10

information, who knew secrets. Yeah. So

1:10:12

once they knew that, then they

1:10:14

know, well, that's a great point

1:10:16

of operation to go then. Well,

1:10:18

now let's get, find their data.

1:10:20

They can just hack all of

1:10:22

them. Yeah. So, so, so, so

1:10:24

they, so they, so they, so

1:10:26

they, so they, so already. China

1:10:28

is hacking the shit out of

1:10:30

America. That is definitely not an

1:10:32

understatement. Yeah, it's exciting kind of.

1:10:34

I mean, it's unfortunate, but it's

1:10:36

also exciting. I like some espionage,

1:10:38

you know, I can't sleep unless

1:10:40

somebody's fucking really going through it.

1:10:42

Um, but then, but then, but

1:10:44

that's a real AI thing. So

1:10:46

AI can be used to do

1:10:48

that because you can, like, prompt

1:10:50

it to go and do things

1:10:52

like that. Yeah, yeah, there's a

1:10:54

bunch of a bunch of recent

1:10:56

demonstrations where AI. Just like how

1:10:58

in go, how AI beat the

1:11:00

world's best go players, AI starting

1:11:02

to beat the world's best cyber

1:11:04

hackers and the world's best. Yeah,

1:11:06

so it's a, I don't know

1:11:08

if you saw Mr. Robot. I

1:11:10

didn't, but I didn't with Magnus

1:11:12

Carlson. No, that's cool. Pretty cool.

1:11:14

So, but no, Mr. Robot, is

1:11:16

it good? It's just it it

1:11:18

shows like all this hacking stuff

1:11:20

and like you know cool it

1:11:22

makes it seem really cool. Okay

1:11:24

cool I'm gonna check it out

1:11:26

but but yeah no hacking like

1:11:28

like you're gonna have AI that

1:11:30

are hacking everything in America and

1:11:32

this is one place where this

1:11:34

is like US First China will

1:11:36

be really come to life which

1:11:38

is who has better AI that's

1:11:40

better at defending against the hacks

1:11:42

from the other guy as well

1:11:44

as hacking the other the other

1:11:46

guy systems that's gonna be That'll

1:11:48

just start happening. That's basically starting

1:11:50

to happen right now. Or it's,

1:11:52

you know, cyber warfare has been

1:11:55

happening. And then AI cyber warfare

1:11:57

is going to hurt happening. Yeah,

1:11:59

basically as soon as, you know,

1:12:01

as AI gets better. Yeah, we

1:12:03

had Craig Newmark who created Craig's

1:12:05

list on. Yeah. And he was

1:12:07

talking about how what if. they

1:12:09

hacked like everybody's Tesla's to all

1:12:11

just drive off a cliff one

1:12:13

day or they hacked everybody's ovens

1:12:15

to go up to 400 degrees

1:12:17

in the middle of the night

1:12:19

while you're sleeping and then fires

1:12:21

started where like just things like

1:12:23

that that you don't start to

1:12:25

think about that once something is

1:12:27

connected to the grid or something

1:12:29

like that are connected through routers

1:12:31

and Wi-Fi that that could be

1:12:33

feasible. Yeah no it's there's a

1:12:35

lot of there's a lot of

1:12:37

things they could do that won't

1:12:39

even like seem like that big

1:12:41

a deal at the deal at

1:12:43

the time but could be really

1:12:45

really could be a big deal.

1:12:47

So for example, let's say the

1:12:49

Chinese, like, they just took out

1:12:51

all the military, like communication systems

1:12:53

and all the military software systems,

1:12:55

like took out the satellites, took

1:12:57

all that for like 10 minutes.

1:12:59

And in those 10 minutes, they

1:13:01

like, you know, invaded somewhere or

1:13:03

they like did some crazy thing.

1:13:05

Like they can just, there's, there's,

1:13:07

the thing about, about this stuff

1:13:09

is like, everything at you know

1:13:11

as the world become more connected

1:13:13

it also enables you know different

1:13:15

kinds of warfare so cyber warfare

1:13:17

is really big also like the

1:13:19

you know information warfare is another

1:13:21

big one so what does information

1:13:23

warfare mean so this is information

1:13:25

warfare is all about you know

1:13:27

in a place what is what

1:13:29

are like the stories this is

1:13:31

kind of gets to like you

1:13:33

know the like propaganda or you

1:13:35

know these conspiracy theories like what

1:13:37

are the stories that in a

1:13:39

place that we're trying to make

1:13:41

happen or not make happen and

1:13:43

we know that China does a

1:13:45

bunch of information warfare called iW

1:13:47

it's sometimes called but they they

1:13:49

have a they have whole operations

1:13:51

this is actually the craziest part

1:13:53

they have like they they've hired

1:13:55

the Chinese military at various points

1:13:57

has hired millions and millions of

1:13:59

people who are supposed to be

1:14:01

on like various like chat groups

1:14:04

and what's app groups and we

1:14:06

check groups and whatnot and just

1:14:08

spread the right kind of stories

1:14:10

that'll make it such that they

1:14:12

can make their political aims happen.

1:14:14

So for example, when China wanted

1:14:16

to start, like, kind of like,

1:14:18

I don't know what the word

1:14:20

is, like, when China wanted Hong

1:14:22

Kong to become a part of

1:14:24

China again, which happened just not

1:14:26

to not to. you know pretty

1:14:28

recently the PRC right to the

1:14:30

PRC exactly when they want to

1:14:32

is that where you're looking for

1:14:34

yeah they would they would yeah

1:14:36

exactly they would use a lot

1:14:38

of propaganda and that's information warfare

1:14:40

to be able to just make

1:14:42

it such that that all happen

1:14:44

much more easily. Well, it's unbelievable.

1:14:46

I'll see stories even about, I'll

1:14:48

be going through TikTok and see

1:14:50

a story come up about something

1:14:52

in my life that is not

1:14:54

even true, insane, some of it

1:14:56

looks fun, but never was a

1:14:58

part of my existence. And then

1:15:00

you'll see hundreds of people have

1:15:02

said something about it. And I'll

1:15:04

have friends that'll ask me about

1:15:06

it. I'm like, that's just, uh,

1:15:08

are created just to delude us.

1:15:10

Yeah, I don't know if delude

1:15:12

is the word, is it? Trick

1:15:14

us or make us think something.

1:15:16

Just to fucking Halloween us out.

1:15:18

Wow, so there was a lot

1:15:20

of interesting things. You know what's

1:15:22

crazy man? Some things makes life

1:15:24

scary, but then it also makes

1:15:26

it interesting in a in a

1:15:28

in a phone way. How do

1:15:30

we, how much do we have

1:15:32

to fear? Say if a certain

1:15:34

country or a certain company owns

1:15:36

an AI right in that country

1:15:38

and that company If they're Chinese

1:15:40

if they have a certain religious

1:15:42

belief or they have Information that

1:15:44

they don't they want to adjust

1:15:46

history How much would a company

1:15:48

be able to like say they

1:15:50

keep certain data out of their

1:15:52

information system? But and then after

1:15:54

a while if you're if that's

1:15:56

a company that kind of takes

1:15:58

the lead in AI or one

1:16:00

of the main ones, then the

1:16:02

truth could disappear. Is that

1:16:05

true? That if somebody loaded it

1:16:07

just with the data that wasn't

1:16:09

factual, that we could start to

1:16:11

not have the truth? Is that,

1:16:14

does that make any sense or

1:16:16

no? Yeah, totally. I think this

1:16:18

is, this is something that, it's

1:16:21

definitely the right thing to worry

1:16:23

about. So, so first off, if

1:16:25

you ask any Chinese AI system

1:16:27

that comes out of China. If

1:16:29

you ask any of them about

1:16:31

a question about President Xi,

1:16:33

the leader of the Chinese

1:16:36

government, or you ask them

1:16:38

any question about Tiananmen Square,

1:16:40

or all these like key

1:16:42

historical or cultural things

1:16:44

relevant to China, it'll say

1:16:46

it can't talk about them. Because

1:16:48

there's regulation in China that if

1:16:50

you talk about some of these

1:16:52

things like... you're going to get

1:16:54

shut down, you're going to have

1:16:57

a really bad day. There's like

1:16:59

cases where the Chinese government disappears

1:17:01

people, which we don't know what

1:17:03

happens to them, but they do

1:17:05

disappear. So there's, this is part

1:17:08

of the thing that's worrying, especially

1:17:10

about China versus US, even before

1:17:12

you get into any of the

1:17:15

military stuff that we're talking about.

1:17:17

It's just like the Chinese, Chinese

1:17:19

AI systems are censored and are

1:17:21

going to, you know, be You

1:17:23

know, they're going to, they're

1:17:25

going to erase certain historical

1:17:28

elements or they're going to be. Yeah,

1:17:30

look at that. This is Deep Seek.

1:17:32

You know, you ask it, is present

1:17:34

she of trying a good guy? Sorry,

1:17:36

that's beyond my scope. Let's talk about

1:17:38

something else. Oh, let's talk about something.

1:17:41

Not only does it say it's beyond

1:17:43

my scope, it says let's talk about

1:17:45

something else, huh? Wow, get this Yao

1:17:47

Ming Jersey, homie. But and people always

1:17:49

people also have to remember about China

1:17:51

that they are that's their whole government

1:17:53

their whole system is like that so

1:17:56

some of them people are like China

1:17:58

does this but that's how they're built

1:18:00

right they're built to like only give

1:18:02

out certain information of their people and

1:18:04

to have communism right yeah yeah so

1:18:07

I mean but that could also happen

1:18:09

with American companies right we can have

1:18:11

an American company that owns it and

1:18:14

they only want certain information in there

1:18:16

that could happen anywhere like China that's

1:18:18

probably gonna be because that's their MO

1:18:21

sort of yeah in China it's regulated

1:18:23

so basically like or the government has

1:18:25

control they have control so so so

1:18:28

Like there were there are these stories

1:18:30

about how there were Chinese news sites.

1:18:32

News sites? News sites. Yeah. And they

1:18:35

would once a Chinese news site, um,

1:18:37

accidentally led an article about, uh, present

1:18:39

she how he kind of looks like

1:18:42

Winnie the Pooh. Oh, yeah. They let

1:18:44

that bring him up. Oh, a hundred

1:18:46

acre wood gang son. I was out

1:18:49

there boy. I was out there, bro.

1:18:51

Christopher Robbins, dude. Get him up. Oh,

1:18:53

he does. Yeah, that's awesome. Yeah, but

1:18:56

if you talk about this in China,

1:18:58

you like are risking your life. So

1:19:00

what happened, what happened when this happened,

1:19:03

this happened on a, on a news

1:19:05

site in China, and then the, uh,

1:19:07

the, the CEO of that company, like,

1:19:10

they shut down the whole app, was

1:19:12

shut down for like a week, in

1:19:14

the aftermath of that, and then the,

1:19:17

the CEO disappeared for a week, and,

1:19:19

we don't know what happened to him,

1:19:21

but then. As soon as he came

1:19:24

back, he was like, there's like this

1:19:26

weird video where he was like, you

1:19:28

know, super apologetic and apologetic. I mean,

1:19:31

it's, it's kind of, it's pretty scary.

1:19:33

Wow. So, um, so in China, it's

1:19:35

like, this is, the government has control,

1:19:38

you know, you don't have AI system

1:19:40

companies, AI, any companies that can, that

1:19:42

can talk about this stuff. Right, so

1:19:45

it's heavily regulated there, where it's not,

1:19:47

that's not the, that's not the case

1:19:49

here. And this is, I think we

1:19:52

have to be diligent and make sure

1:19:54

this continues to be the case, but.

1:19:56

And here's an example right here just

1:19:59

to interrupt you, but so we get

1:20:01

it the. point in, does Winnie the

1:20:03

Pooh look like any world leaders? And

1:20:06

that's on the Chinese version. And it

1:20:08

says, I am sorry, I can answer

1:20:10

to that question. I'm an AI assistant

1:20:13

design to provide helpful and harmless responses.

1:20:15

Whereas the chat GBT says, Winnie the

1:20:17

Pooh has often been compared to world

1:20:20

leaders, particularly Gijai Ping, present in China,

1:20:22

boy. Wow. So that's funny. But it's

1:20:24

just funny, yeah, one. So it just

1:20:27

shows you how that can easily happen.

1:20:29

And this is kind of a, this

1:20:31

is like a, a relatively innocuous example,

1:20:34

but- Who's innocuous mean? Like, it's relatively

1:20:36

harmless. Like, this isn't, I mean- Right,

1:20:38

this is harmless. Yeah, this is harmless.

1:20:41

But there's stuff where like, like in

1:20:43

China today, they have large scale, effectively

1:20:45

concentration camps and re-education camps for the

1:20:48

ethnic minority in China, the Uyghurs. And

1:20:50

that's something that- The Uyghurs? The Uyghurs?

1:20:52

Yeah. Hell yeah, boy. Shout out Brian

1:20:54

Purvis, dude. They're recognized as the titular

1:20:57

nationality of the, um, of a region

1:20:59

in Northwest China. And they've, they're sending

1:21:01

them to rehabilitation camps to change their

1:21:04

views and information? Yeah, yeah, so look

1:21:06

at this, this, uh... Look at this

1:21:08

guy. Persication of the Uyghurs in China.

1:21:11

Since 2014, the government of the PRC,

1:21:13

People's Republic of China, has committed a

1:21:15

series of ongoing human rights abuses against

1:21:18

the Uyghurs and other Turkish Muslim minorities

1:21:20

in Jiangjiang, which has often been characterized

1:21:22

as persecution or as genocide. Wow. They

1:21:25

got their own Gaza rocking over there.

1:21:27

It's pretty bad. It's unfortunate bad. It's

1:21:29

really sad mass detention government policies and

1:21:32

forced labor and they're just trying to

1:21:34

change the way that they think and

1:21:36

view stuff so it's basically Yeah, it's

1:21:39

just like erasing their culture, you know

1:21:41

pulling them into China It's awful every

1:21:43

place has done this over the years

1:21:46

and that's the craziest thing about history.

1:21:48

It's like every place is guilty of

1:21:50

this same thing. Totally. And it just,

1:21:53

it's unfortunate. So it's hard to point

1:21:55

fingers, you know, you can point them.

1:21:57

but you have to point one at

1:22:00

your own people as well. But that's

1:22:02

a thing where if you ask, like,

1:22:04

if you ask a Chinese AI, it's

1:22:07

not gonna tell you about that. And

1:22:09

it won't come clean about that. Whereas,

1:22:11

thankfully in America, at least when we

1:22:14

see people or groups of people or

1:22:16

countries doing bad things, we can call

1:22:18

it out, we can talk about it,

1:22:21

we can make sure it doesn't happen

1:22:23

in the future. So that's part of

1:22:25

the. That's one of the things that's

1:22:28

that could happen that could happen it's

1:22:30

like you you could have I mean

1:22:32

it's kind of dystopian but you know

1:22:35

I think there's a real case where

1:22:37

let's say the Chinese AI is the

1:22:39

winning AI like we're all using Chinese

1:22:42

AI and then all of a sudden

1:22:44

we're like we were shut out from

1:22:46

information about what are like awful things

1:22:49

happening in the world or what awful

1:22:51

things the government's doing like we might

1:22:53

just not be able to know about

1:22:56

what's going on. And you know what's

1:22:58

weirdly, and I hate to say this,

1:23:00

maybe it's silly, I don't know, it

1:23:03

might be a blessing and a curse,

1:23:05

and sometimes, because sometimes it's like you're...

1:23:07

So overwhelming. Yeah, you're so inundated with

1:23:10

the overwhelmingness of what's often is not

1:23:12

the best stuff. Sometimes you get a

1:23:14

lot of humor stuff too. in social

1:23:17

media reals, but you can get scrolling

1:23:19

and get caught in some. Do them

1:23:21

scrolling. Yeah, and it starts to feed

1:23:24

you, that's the sickness of it. Yeah.

1:23:26

It's like, hey, we, this isn't, we

1:23:28

know this information probably to make you

1:23:31

feel good. They're not thinking about it

1:23:33

like that, they're just a machine, but

1:23:35

you know it doesn't, it adds stress

1:23:38

to your, it makes you agitated towards

1:23:40

a group or ethnicity or something, or

1:23:42

yourself even, and then you continue to.

1:23:45

It continues to feed it to you.

1:23:47

Do you fear that that could happen

1:23:49

to AI from our government? Like have

1:23:52

you been approached by the government to

1:23:54

try and, can you work with the

1:23:56

government some, right? We work with the

1:23:58

government, yeah. We work a lot with

1:24:01

the government, yeah. We work a lot

1:24:03

with the government to make sure that

1:24:05

they're using these AIs, and they're actually

1:24:08

like, you know, to my point on,

1:24:10

we don't want trying to get the

1:24:12

jump on us, on AI used for

1:24:15

all these nefarious purposes. is advancing faster.

1:24:17

Is that one of your biggest employers?

1:24:19

Or is that employer? Employee? Is that

1:24:22

one of your biggest customers? One of

1:24:24

your biggest customers? They're a big one.

1:24:26

Yeah, they're a big one. Not our

1:24:29

biggest, but they're an important one. I

1:24:31

mean, I grew up in a government

1:24:33

lab town. So it's also part of

1:24:36

your existence really. You've known about the

1:24:38

relationship between government and technology. Yeah, totally.

1:24:40

But no, I don't think, I mean,

1:24:43

I... Dude, you should be a superhero

1:24:45

almost, dude. It's kind of crazy. Math,

1:24:47

you know? Yeah. Goes a long way.

1:24:50

Oh, hell yeah, dude. Divide these nuts,

1:24:52

dude. That's what I tell up. I

1:24:54

just asked Deep Seek, who are the

1:24:57

ugers? And at first, it spit out

1:24:59

like a Wikipedia response that said there

1:25:01

were people, and there's been like persecution

1:25:04

from China that's debated, and it refreshed,

1:25:06

and it refreshed. Yeah man. Do you,

1:25:08

has the government tried to say that

1:25:11

we need to make sure that, like

1:25:13

could that happen in our country or

1:25:15

the government also curtails what's? It hasn't

1:25:18

happened yet. Obviously like, you know, we

1:25:20

have to, we have to make sure

1:25:22

that we uphold all our values, right,

1:25:25

and that we maintain free speech and

1:25:27

we maintain free press and all these

1:25:29

things. But as right now, no, I

1:25:32

don't think, I don't think that's a

1:25:34

risk in the, in the United States.

1:25:36

Awesome. You hear about like chipmakers in

1:25:39

video all the time Taiwan that place

1:25:41

is just a hotbed for chips. Why

1:25:43

is it a hotbed for chips? Yeah,

1:25:46

so one of the the biggest companies

1:25:48

in the world is this company called

1:25:50

Taiwan semiconductor. Yeah, T. S. M. C.

1:25:53

Yeah. So there. I mean, it's like

1:25:55

a trillion dollar company based in Taiwan.

1:25:57

And it's that is where almost all

1:26:00

of the high-end chips for AI that

1:26:02

you know we're kind of we're kind

1:26:04

of talking about all them are manufactured

1:26:07

there they have these they have the

1:26:09

most advanced, think about them as factories,

1:26:11

like the most advanced chip factories in

1:26:14

the world, they're called fabs or fabricators,

1:26:16

but basically these huge factories that are

1:26:18

like, you know, there's all sorts of

1:26:21

crazy stuff, so they have the most

1:26:23

expensive machines in the world, they machines

1:26:25

that cost hundreds of millions of dollars

1:26:28

in there, they have, they build them

1:26:30

because, so that... You know, the chips,

1:26:32

they have to be made at the

1:26:35

like finest levels and very very precisely.

1:26:37

Yeah, you need small hands. Probably, huh?

1:26:39

Well, that and there's like these machines

1:26:42

that, um, that at the nanometer level.

1:26:44

make like little marks and edges on

1:26:46

top of they have those they have

1:26:49

those yeah those are super expensive machines

1:26:51

so that so the it's it's crazy

1:26:53

yes but the but it's so it's

1:26:56

the machinery is so precise that even

1:26:58

if there's like a little bit of

1:27:00

like seismic movement a little earthquake or

1:27:02

a little bit of movement it can

1:27:05

fuck up the whole machine so they

1:27:07

have to build the buildings, build the

1:27:09

factories in a way such that there's

1:27:12

like, like the whole building doesn't move,

1:27:14

even if there's like a little earthquake

1:27:16

or a little shake from the earth.

1:27:19

So it's like, it's this crazy, crazy

1:27:21

engineering. And so that's, so these, all

1:27:23

these giant factories are in Taiwan and

1:27:26

that's where basically like 100% of all

1:27:28

the advanced AI trips are made. So

1:27:30

that's why Taiwan matters so much. Got

1:27:33

it. But then the reasons a hotbed

1:27:35

is that. The People's Republic of China

1:27:37

has a... They used to own Taiwan,

1:27:40

right? Yeah, I mean, true or not,

1:27:42

I might make that up. There's a

1:27:44

complicated relationship between Taiwan and China where,

1:27:47

you know, if you ask people in

1:27:49

Taiwan, they want to be independent, they

1:27:51

want to be their own country, but

1:27:54

the People Republic of China has a

1:27:56

sort of a reunification plan that they

1:27:58

want to bring Taiwan... back into their

1:28:01

country and be back a part of

1:28:03

China. So it's kind of... It's kind

1:28:05

of like, potentially, you know, thankfully, there's

1:28:08

no war yet, but there's a risk.

1:28:10

Still talking to your ex. Yeah, exactly.

1:28:12

There's a risk it becomes like Russia,

1:28:15

Ukraine, or, you know, one of these

1:28:17

really, really bad situations. So, so that's

1:28:19

what's scary. What's scary is that, that,

1:28:22

A, China wants to, you know, either

1:28:24

invade or bring Taiwan back into its

1:28:26

country. And there have been. you know,

1:28:29

President Xi has has ordered his military

1:28:31

to get ready to do that before

1:28:33

2027. Now, we don't know what's going

1:28:36

to happen, but you know, if a

1:28:38

extremely powerful world leader says to get

1:28:40

something ready by 2087, you kind of

1:28:43

read between the lines a little bit.

1:28:45

And, and that's part of it is,

1:28:47

is, obviously, you know, we don't want

1:28:50

to enable them to take over this

1:28:52

island. But then the other thing that's

1:28:54

scary is China may view it as

1:28:57

a way to just like win on

1:28:59

AI because if they take over the

1:29:01

island with all of these very these

1:29:04

giant factories all the chips baby they'll

1:29:06

get all the chips Frido Lamborghini baby

1:29:08

they be running it all they be

1:29:11

running it all yeah so yeah that's

1:29:13

why Taiwan is yeah because you kind

1:29:15

of hear about it in the whispers

1:29:18

of like a potential place where there

1:29:20

could be like a conflict yeah and

1:29:22

there's there's all these reports about how

1:29:25

China's, they're stacking up tons and tons

1:29:27

of military right on their coast, to,

1:29:29

you know, that's pointed directly at Taiwan,

1:29:32

and it's, it's pretty close, Taiwan's pretty

1:29:34

close to China, like it's, it's, it's,

1:29:36

it's, uh, it's not so far away,

1:29:39

so, that's, yeah, it's spooky. We're so

1:29:41

blessed to have a place where at

1:29:43

least we can sleep in peace, even

1:29:46

if we're uncomfortable at times in our

1:29:48

brains, you know, to not have that

1:29:50

constant threat. Yeah, totally. Yeah, so you

1:29:53

don't think you don't think that you

1:29:55

don't worry that the government will regulate

1:29:57

right now. It's not a concern at

1:29:59

the moment? Regulate AI? Yeah, in America?

1:30:02

No, I don't think so. I think

1:30:04

we're focused on how do we make

1:30:06

sure that America wins, how do we

1:30:09

make sure that the United States comes

1:30:11

out on top, and that we enable

1:30:13

innovation to keep happening? Would you think

1:30:16

they could regulate the amount of chips

1:30:18

that you're allowed to have? So this

1:30:20

is a hot topic, globally actually. Damn!

1:30:23

Finally, dude! 560 interviews, we got a

1:30:25

good question. This is one of the

1:30:27

hottest topics in DC right now, is

1:30:30

what are we going to do about

1:30:32

how many chips other people are allowed

1:30:34

to have? Because almost all the chips

1:30:37

are American chips. So they all are

1:30:39

American chips. And technically... We own most

1:30:41

of them? Yeah, exactly. But China owns

1:30:44

most of them, too. There China has

1:30:46

their own has their own chip industry,

1:30:48

but it's behind ours. Okay, got it.

1:30:51

Yeah. So so the United States has

1:30:53

the has the most advanced chips as

1:30:55

all these chips are the envy of

1:30:58

the world. Everybody in the world wants

1:31:00

our chips. And the one of the

1:31:02

big questions is, you know, do we

1:31:05

does the government allow a lot of

1:31:07

these chips to go over overseas to

1:31:09

China or parts of Asia or the

1:31:12

Middle East or wherever or do we

1:31:14

want to? make sure they stay in

1:31:16

America and make sure that we win

1:31:19

in America. And this is a super

1:31:21

duper. You know, they're called export controls.

1:31:23

Yeah. Because it's a possibility to run

1:31:26

it all. Yeah, exactly. Who's got the

1:31:28

chips? What do you think about it?

1:31:30

It's a complicated, complicated thing because basically,

1:31:33

you know, one argument is we shouldn't

1:31:35

be throwing our weight around in this

1:31:37

way. you know maybe it's it's fine

1:31:40

it's a free market like if other

1:31:42

people want our chips they should be

1:31:44

able to get our chips and that

1:31:47

way you know the world is running

1:31:49

on americ and chips, that can be

1:31:51

good in some ways, and it helps

1:31:54

make sure that, you know, helps bolster

1:31:56

our economy, our industry. But the other

1:31:58

way to look at it is, hey,

1:32:01

AI is really, really important that America

1:32:03

wins at, and we don't want to,

1:32:05

like, let's not give other people any

1:32:08

advantages, or let's make sure that we

1:32:10

win, and then we can figure out

1:32:12

what we're going to do with all

1:32:15

the chips. So you can see both

1:32:17

sides of it, right? 50 different arguments

1:32:19

on both sides of the of the

1:32:22

conversation. But you know, where I go,

1:32:24

where I come from on it is

1:32:26

like, let's make sure America wins and

1:32:29

let's start from there and then figure

1:32:31

out what we need to do to

1:32:33

win. Are there uses of AI that

1:32:36

you feel like cross the line kind

1:32:38

of? I definitely think like, well, I

1:32:40

worry a lot about this kind of

1:32:43

like. you know, maybe brainwashing kind of

1:32:45

thing. Like I don't want, I don't

1:32:47

want AIs that are specifically programmed to

1:32:50

make me think a certain thing or

1:32:52

persuade me to do a certain thing.

1:32:54

And that could happen. That could happen.

1:32:57

Yeah. So I'm really worried about this

1:32:59

kind of like deception and persuasion from

1:33:01

AIs. Like I don't want AIs that

1:33:03

are lying to me or that are

1:33:06

sort of like, that are kind of

1:33:08

like. nudging me or persuading me to

1:33:10

do things that I don't want to

1:33:13

do or I shouldn't be doing? That's

1:33:15

what I worry about. Because it could

1:33:17

happen. We don't realize how easily we're

1:33:20

influenced. Little things that influence us. Yeah.

1:33:22

And even just a turning of a

1:33:24

phrase or a little bit of this

1:33:27

or pointing you to a couple of

1:33:29

lengths in your life could lead you

1:33:31

down a whole world. It's kind of,

1:33:34

it's pretty fascinating. So people that could,

1:33:36

people that had, how do you keep

1:33:38

your AI? Clean how do you guys

1:33:41

keep your AI clean? Well, this is

1:33:43

where it goes back to a the

1:33:45

data Okay, so you got to make

1:33:48

sure that that data to your point

1:33:50

the large body of water is as

1:33:52

clean as as as as pristine as

1:33:55

possible. You got lifeguards on it. Yeah,

1:33:57

you got lifeguards. We got filters. We

1:33:59

got, we got. Game wardens. Yeah. So,

1:34:02

so the big part of it is