0:00

The hack was one of the

0:03

pivotal moments of Bitfinex. That

0:06

was incredible. In

0:09

seven months, I think by April

0:11

2017, all the outstanding, the

0:13

effects tokens were either converted,

0:16

redeemed, or converted in equity.

0:19

The really simple idea around Tether

0:22

was, okay, let's reuse

0:24

the incredible technology that

0:26

Bitcoin created called blockchain.

0:29

Let's just put the US dollar

0:31

because it's the most used currency in the world. I

0:34

always say that before 2022,

0:36

we never had even a

0:38

marketing team for Tether. So

0:42

Tether USDT has become really

0:45

important for hundreds of millions of people

0:47

across the board, especially in emerging markets

0:49

and developing countries. This

1:05

episode is brought to you by Gnosis. Gnosis

1:07

builds decentralized infrastructure for the

1:10

Ethereum ecosystem. With a

1:12

rich history dating back to 2015 and

1:15

products like Safe, Cowswap, or Gnosis

1:17

Chain, Gnosis combines

1:19

needs-driven development with

1:21

deep technical expertise. This

1:24

year marks the launch of Gnosis

1:26

Pay, the world's first decentralized payment

1:28

network. With a

1:31

Gnosis card, you can spend self-custody

1:33

crypto at any Visa accepting merchant

1:35

around the world. If

1:37

you're an individual looking to live more

1:39

on-chain or a business looking to white

1:41

label the stack, visit

1:44

gnosispay.com. There

1:46

are lots of ways you can join the Gnosis

1:48

journey. Drop in the Gnosis

1:50

DAO governance form, become a

1:52

Gnosis validator with a single GNO

1:54

token and low-cost hardware, or

1:57

deploy your product on the EVM-compatible

1:59

and highly decentralized Gnosis chain.

2:02

Get started today at gnosis.io.

2:07

Chorus One is one of the biggest node

2:09

operators globally and help you stake your tokens

2:11

on 45 plus networks

2:13

like Ethereum, Cosmos, Celestia

2:15

and DYDX. More

2:18

than 100,000 delegators stake with Chorus

2:20

One including institutions like BitGo and

2:23

Ledger. Staking with Chorus

2:25

One not only gets you the highest years,

2:28

but also the most

2:30

robust security practices and infrastructure

2:32

that are usually exclusive for

2:34

institutions. You can

2:36

stake directly to Chorus One's public node

2:38

from your wallet, set up a wide

2:40

table node or use the recently launched

2:42

product Opus to stake up to 8,000

2:44

ETH in a single transaction.

2:47

You can even offer high-year staking

2:50

to your own customers using

2:52

their API. Your assets always remain

2:54

in your custody so you can have

2:56

complete peace of mind. Start staking today at

2:59

chorus.one. Welcome

3:02

to App Center, the show which talks about

3:04

the technologies, projects and people driving decentralization

3:06

and the blockchain revolution. I'm Brian Crane and

3:08

I'm today here with Frederik Erenst. Today

3:11

we're speaking with Paulo Ardore-Noff who is

3:13

the CEO and co-founder of Tether and

3:16

CTO of Bitfinex and also working on

3:18

another new thing called Holepunch which we'll

3:20

talk about. So of course some

3:22

enormously influential projects in the crypto

3:25

space. So this

3:27

Tether is really excited to get into this.

3:30

Paulo, thanks so much for coming on. It's really

3:32

great to have you. Thank

3:35

you for having me. We're

3:37

getting right into it. You've been in crypto

3:39

for a really long time and have worked

3:41

on many different things. But how did

3:43

it all start? How did you become

3:46

interested in crypto and what triggered your

3:48

imagination? So,

3:51

2024 is my 10th

3:54

year in business in

3:56

crypto. I discovered Bitcoin

3:58

in 2013. So,

4:01

I've been a developer all my life. Now

4:04

I'm focusing more on business and

4:07

strategy, but I was

4:09

born as a developer since I was eight

4:11

years old, I was coding. I

4:14

was living in a really small town outside of the

4:16

city and didn't have many

4:19

friends around. So I spent almost

4:21

all of my afternoons with a

4:23

computer. So fast forward, went

4:25

to the university in Genoa, in

4:27

Italy, and studied math applied

4:29

computer science. And

4:34

I'm also a big sci-fi fan. So

4:37

my dream and

4:39

my interest in technology was all

4:41

about building things, building

4:44

technological solutions that would be resistant

4:48

to apocalypse. So

4:50

in the house sci-fi, there's always

4:52

problems for the world, things getting

4:54

for, going for the worst. You

4:58

have either wars or you have aliens or

5:00

you have to abandon the earth and go

5:02

somewhere else and so on. So that

5:05

thing always was

5:07

sticking to my mind on how whatever

5:10

we do, whatever we build should not

5:12

be built for the best case scenario,

5:14

but should be built for the worst

5:16

case scenario. And so

5:19

I started, as soon

5:21

as I graduated from the university, I got

5:24

one of the most interesting jobs ever

5:26

as a researcher at the university. So

5:29

I was building resilient

5:31

telecommunication systems for

5:33

distressed situations like

5:35

battlefields. And

5:37

that taught me how you can

5:40

build really resilient networks

5:42

that are really critical and

5:44

crucial in certain moments. Then

5:47

as most of the time happens, in

5:49

Italy, you don't get paid too much.

5:53

So that's the

5:55

unfortunately situation Of

5:57

Italy. So I Decided to move to a field.

6:00

The older data I I thought was

6:02

more rewarding at least in economical pay.

6:04

that was finance. So I got my

6:06

first job in financing the in utterly

6:09

Switzerland and then I decided to move

6:11

a to London to open my start

6:13

up to be a Difference a software

6:15

So in two thousand twelve I opened

6:18

the start up and the in two

6:20

thousand and thirteen I I was. Already.

6:24

Read annoyed. By. The

6:26

outdated technology that the financial system

6:28

was rely on. If

6:30

you haven't worked with the. Bangs.

6:32

Or has plans as saw You know

6:34

that every time. They trade

6:36

any time a trade or trades if for

6:39

an institution of wear headphones, Yeah.

6:41

The added the day you have to

6:43

reconcile old positions, all the trades, all

6:45

the cash balances across multiple trading venues.

6:48

Across. Multiple banks and custodians,

6:51

And. All that is splitting

6:53

the tens of different protocols is

6:55

kept together by rubber bands is

6:57

all. Really outdated. Feels

7:00

like is all steel returning. Cobbled.

7:03

Forty years ago. So in two

7:05

thousand and thirteen I go to

7:07

the chance to learn about Bitcoin.

7:10

And. The first

7:12

thing that has thought about bitcoin was okay

7:14

this is really cool. Disease.

7:16

The not not just, I

7:18

didn't get immediately. Honestly, the.

7:21

Humanitarian. A social. Implications.

7:25

But. I. Got

7:27

the technology aspect and I thought

7:30

that bitcoin could be. The

7:33

solution for. The. Financial

7:35

Board in terms of having a common

7:37

settlement network across old is he do

7:39

since across all the government's the beautiful

7:41

thing about Bitcoin as above trainees that

7:43

you can see. Everyone

7:46

had that has a synth node sees

7:48

the same thing. great so you you

7:50

cannot have ever seen reconciliation and so

7:52

on. So is beautiful, prevents double spending,

7:54

gun and and so on. so I

7:56

was really excited about it and then

7:58

or the other to doesn't there. The

8:00

and i thought really i was really d

8:02

been them The rabbit. Hole. And

8:04

so I started understanding more.

8:07

How. Bitcoin is actually game changer

8:09

as a currency for and

8:11

it's social impact could be

8:13

incredible. So. Long.

8:16

Story short: in two thousand and fourteen I

8:18

got the chance of i'm. Of

8:20

the war team for between Act.

8:23

So Bitterness was one of the

8:25

first trading platforms. Was. Really

8:27

one of probably the only and first

8:29

trading platform that at the time was

8:31

offering marching trading. On. Bitcoin.

8:34

And so the see a soldier was

8:37

any of the beat. The next contacted

8:39

me. Through. Com um friend and

8:41

we as he asked my help in

8:43

in making sure that bitterness could scale

8:45

with a demand right to Doesn't Fourteen

8:48

is not, doesn't Twenty four of course

8:50

we're seeing today the this you to

8:52

the mantle of Bitcoin. In

8:55

a burrow it's by a by

8:57

a d ass and so on.

8:59

But in two thousand and fourteen

9:01

sealed there was an increase in

9:03

demand for for for accessing between

9:05

trading. As. So. Most.

9:07

Of the it seems he's in two thousand

9:09

and fourteen were. Built. By.

9:13

I'm a Tories no threats. Hermes is or

9:15

as it is not about you know been

9:17

A but we're more mostly almost. Ecommerce is

9:19

for between rather than high performers trading platforms.

9:21

Today we are used to have super Alone

9:23

A to the I frequency trading and saw

9:26

him but back in time was kind of

9:28

different. So. I brought my expertise

9:30

in in building in building the you

9:32

know the beats, an expat form and

9:34

the setting up a team. In two

9:36

thousand and sixteen I became the city

9:38

or bit Phoenix and the in the

9:40

two thousand and seventeen I became the

9:42

sea is city or of are also

9:44

Tatar. And. Then from there

9:47

I started working also on the

9:49

stable coincide. For

9:51

tatar and them more he simply

9:53

I started to care. More.

9:56

About term long term strategy,

9:58

a Visa and. Their

10:00

growth though the teams and execution of

10:02

them. the strategy for for that are

10:04

was feel I'm the seat your for

10:06

beating so I would take care of

10:08

the taskbar form and they're all the

10:10

trading tools build around beat the next.

10:13

So. Did his just means that than

10:15

five minutes? Probably more. But yeah no

10:17

no thank you. So us for ah

10:19

for dad am curious. Like

10:21

going a little they didn't sort

10:23

of the story of between eggs

10:25

what do you think it was

10:27

Dad helped. Between x

10:30

and. Succeeds. And

10:33

and actually this maybe one episode I'd

10:35

really be curious. the top se goes

10:37

to talk about that thing. Some people

10:39

will still remember advice because once. You.

10:41

Guys have this massive hacked and then yeah,

10:44

it's very interesting. way of like dealing with

10:46

the fact that effing ended up working out

10:48

really well. By. Also love as

10:50

he could sort of talk through.episode and.

10:53

What? Happened. then. I. Think

10:55

this ride a sink. The.

10:57

A hug. Was. One of them

11:00

the people tell moments of bit cynics.

11:03

Remember deaths from the board was

11:05

coming where the be clear ward

11:07

who was coming from. I'm a

11:09

big fact that was the Mongols

11:11

and right so. At

11:13

a time when Bitcoin finance was the

11:15

hands, it was a. Personally,

11:18

Add diversity moment. Ah

11:21

Elsa for of course for the

11:23

company for for our customers. But.

11:26

Look, we had two choices. Either.

11:29

Letting go and let a bit

11:31

phoenix but they city die. Or.

11:35

We. Could. Roll. Up our sleeves,

11:37

And. Make sure that our customers would

11:40

get whole right? So at a

11:42

time when between a thesis was

11:44

hacked was the second of august

11:46

doesn't a sixteen. Up

11:48

because was around six hundred dollars. So.

11:51

the total loss of them have

11:53

beaten it was around seventy two

11:55

million dollars so around one hundred

11:57

nineteen thousand bitcoin so we the

11:59

decided, so I remember those days

12:02

really well because after the

12:04

first day or so of assessment of what happened,

12:06

how it could have happened, so on, we

12:09

decided to come up with a

12:11

plan to resume trading and make

12:13

everyone whole. The way

12:15

that we used was actually

12:18

new and I

12:21

think was extremely intelligent

12:24

and forward thinking.

12:27

So we do not distribute

12:36

the so-called BFX tokens across the users and

12:39

we open the market for these BFX

12:44

tokens so that

12:47

people could trade their

12:51

basically IOUs. The

12:53

interesting thing is that these tokens had multiple

12:55

functions, right? So people could take these tokens

12:57

and convert them into equity if they believe

12:59

that the finance would be successful in long

13:01

term. They could trade them on the

13:04

market, so sell them, for

13:06

example, and they

13:08

could also wait for the

13:10

redemption. So we

13:12

committed to redeeming those tokens

13:15

at $1 face

13:17

value with the revenues of

13:19

the exchange. So

13:21

as soon as the market started, so after

13:23

one week, so it took one week to

13:26

bring back the platform up and running. I

13:28

remember that probably it's left two hours in a week.

13:31

I was completely destroyed after seven

13:33

days, but we had

13:35

to make it. We had to

13:38

put everything we had to

13:40

start trading again, because if you

13:42

wait too much, people will

13:46

just disappear, will go away, and there

13:48

is no chance of success anymore. So

13:51

we knew that we had to restart trading within one

13:53

week. So as soon

13:55

as we started trading, the BFX

13:57

tokens started selling their BFX tokens.

14:00

And the price went from one. Two

14:02

three to since. You.

14:04

Know that they thought, well these guys will

14:07

never make it. After.

14:10

The first days we had, The the first days

14:12

we had a huge outflow of of bitcoin from

14:14

the platform. No. Person of course

14:16

were withdrawing in panic. Buttons.

14:20

After. A few days after ten days

14:22

he's been days. We were back. the first

14:24

exchange by trading volume. So.

14:26

At the end of June by of the entered

14:28

although this we were the first again the biggest

14:30

exchange. By. Three have voted as

14:32

so by the end of September.

14:35

We. Redeem the first effects

14:37

tokens. So people with were

14:39

were like are only cow this guy's these

14:42

guys are making money they are buying back

14:44

the tokens as so the price on the

14:46

secondary market for Dc be effect stock as

14:48

dark to go back and so then October

14:50

team November team we we we kept making

14:52

more money because also desk lose sight of

14:54

Weed de Boer am I the started war

14:56

and by the end of two thousand and

14:58

sixteen going to to doesn't seventeen. That was

15:00

one of the most incredible years for for

15:03

Bitcoin to decrypt have a system. So

15:05

we were making really good profits my back

15:07

if the idea that Saugus a more more

15:09

people stuck to. Convert. Their be

15:12

effect stock and seems to equity so

15:14

that's how basically we got almost three

15:16

on their cheryl their support for for

15:19

I, Phoenix and erm. That.

15:21

Was the incredible I think by

15:23

by the in seven months I

15:25

think by By Gabriel two thousand

15:27

and seventeen all the outstanding. The

15:29

effects tokens were either you know,

15:31

converted, redeem to or or converted

15:33

the name inequity. So it's I

15:35

really. Like. And love to think about

15:37

the that part of the story because I think that the

15:40

him. In these moments of

15:42

weakness is how you really assess the

15:44

team and I must say that no

15:46

one in our team left the company

15:48

at one was focus on make sure

15:50

that so the platform would be profitable

15:52

to make every full. Yet.

15:55

To this day it's one of the

15:57

biggest comebacks at in in crypto history.

16:00

And I also really appreciated. And the sorry

16:02

of the B S x so thin how

16:04

kind of you use good judgment on makes

16:06

it kind of. To. Make it work

16:08

for you guys. So yet at tipping

16:10

my head to you guys. So

16:13

you. Talked about a kind of. It's.

16:15

You know, sitting at the financial records and head

16:17

of settling things. On. Block shame because

16:19

it makes sense and. I can

16:21

say one hundred to sense of and of. We

16:23

have recently started bidding. Sign

16:26

answer and infrastructure that kind

16:28

of crosses the chasm from

16:30

blocked same to Legacy side

16:32

and and with Noses Pay

16:34

and other such integrations. and

16:36

I was absolutely flabbergasted at

16:38

the stack that kind of

16:41

like that. The sign answer

16:43

wide. Runs. On. It's

16:45

it's. unbelievable. So you guys in

16:47

that vein, you guys started and

16:49

eighty cents ice exchange and twenty

16:52

seventeen back when? these were kind

16:54

of them. says. Sadie

16:56

Marginal. So. I'm

16:58

it was easy Next ended later

17:00

rebranded to diversify him and now

17:02

I know Fi. How

17:04

do you jagger that so easily hidden

17:07

in times as fast as news and

17:09

so on? Obesity. His being am a

17:11

decks is preferable to having a sanitized

17:13

exchange in some sense because kind of

17:15

like everything is transparent. Kind of the

17:18

spirit is that. So. How

17:20

do you see the coexistence of both of

17:22

these broads? Bunches? And

17:24

so I think that's the beauty of the of

17:26

Beats the next. And his group data is always.

17:29

You. Know there's this meme of have been here

17:31

before for the attack right? So then you

17:33

know you and the people with the be

17:35

Gov gold teams and then and and sorts

17:38

of but. We have been

17:40

always really. In. In be

17:42

sealed the for for the tac right?

17:44

We always try to push the boundaries

17:46

of of tech even when. Deputies

17:49

were not a thing We thought how we

17:51

could. Make. Sure that

17:53

people. See. people are not

17:55

comfortable leaving their tokens and their money

17:57

on the exchanges we

18:00

can make the experience,

18:02

their user experience great in a

18:04

sense that centralized exchanges

18:07

have the advantage of

18:09

being extremely fast in execution. Because

18:11

the thing about the

18:14

fact that now the average latency of

18:16

an order on Bitfinex is around, in

18:18

the internal system is around 10 microseconds,

18:23

between 10 and 20 microseconds, right? So

18:25

that's how you scale, you can handle any

18:28

sort of volume, and so on. And you

18:30

can only do that through a

18:33

centralized platform. Because even

18:35

the fastest blockchains like Solana, they have

18:38

a block timer for under milliseconds. So

18:40

that is incredibly, incredibly longer. It's like

18:42

an eternity compared to the latency that

18:44

you have in a centralized system. But

18:46

you know, the problem there is just

18:48

something in physics called the speed of

18:50

light, right? The information goes to

18:53

the speed of light. And so you cannot

18:56

move information if you have multiple nodes that

18:58

have to concur to the same information, they

19:00

have to agree, they have to exchange

19:03

information, that speed is the speed

19:05

of light. And so

19:08

as fast as it can be, if you

19:10

have to have two different, two different round trips

19:12

around the world, because one node could be Tokyo,

19:14

one could be in Switzerland, the other one could

19:16

be in the US, then, you know,

19:19

you get to 400, milliseconds

19:21

of a minimum time that's needed for

19:23

a blockchain, and so for

19:25

a DAX to in terms of latency.

19:29

Now, so the way we approach this

19:31

is what are the pros and cons of

19:33

a DAX compared to a centralized

19:35

exchange. And for us, the

19:38

most important part was the having the

19:40

ability of keeping your own

19:42

custody while trading. So

19:45

being able to trade as fast as

19:48

fast as you can with the normal

19:51

centralized user experience, but at the same time,

19:53

keeping control over your own custody. I think

19:55

that is the holy grail of

19:57

of finance, if I have to think it's

20:00

The Room. Course. The the

20:02

other important part. Is.

20:04

The transparency arrays so you want to make

20:06

sure that people them from there are new

20:08

and so on. but. They think that to

20:10

the transparency. Can be.

20:13

On one side of course and dashes you have

20:15

a spicy but you have a normal slaves in

20:18

see so you can also do We had there

20:20

were. Also pays his in which

20:22

people were from run. Because

20:24

the you know the is looking at

20:26

the men cool looking at term you

20:29

know the it'll have even the on

20:31

the divide a date or for example

20:33

cool ten from trans people because they

20:35

are sievert as actions a little bit

20:37

sooner So they're also problems like that

20:40

in the In That race or to

20:42

me the Holy Grail and the Big.

20:45

Advantage of attacks is the ability over of

20:47

are letting people control their own sons. So

20:49

that's why we started as the next in

20:52

that way. As of today, I

20:54

think you know the. There.

20:56

Is another. or there

20:58

was another. interesting. Kind

21:00

of advantage of that sees

21:02

vs centralized the changes. That.

21:05

Is privacy or anonymity if you've

21:07

got a few. will rank. So.

21:10

With that says you usually log into your

21:13

with your met asked. And.

21:15

Are so we'll or whatever other what'd

21:17

you have and then you you can

21:19

start pretty media to you don't have

21:21

to go through he was cml and

21:24

so on. But. Serb my

21:26

opinion as of today seen also

21:28

meet and the other regulations death

21:30

par to death advantage of taxis

21:32

will come to an end. I

21:34

think he is. You know regulators

21:37

make it. Made. It abundantly

21:39

clear. That. The ethos

21:41

you trade against. Someone

21:44

else and that someone else

21:46

ease of person that the

21:48

sanctions. The. Fact that you

21:50

are using attacks is not going to save

21:52

you or pray to justification for you. So

21:55

basically what will it end up in

21:57

my opinion of from what I gather,

22:00

Been. The. Only advantage of

22:02

that sees that is still a great

22:04

advantage is the ability to of keeping

22:06

your on custody of the fans and

22:09

not be subject to put a southern

22:11

accent Hands. So.

22:13

With see the future is will

22:15

will tell us who mental power

22:17

thing this thing with will develop.

22:21

I. Wanted the Elena the ask another question. that

22:23

sort of. Relayed of in a

22:25

bid to those bid for next journey but

22:27

also like more generally it's huge or any

22:29

like as an entrepreneur i mean you mention

22:31

sort of like how. Did seem

22:33

so have made it through all of

22:36

dad and and like. A.

22:38

You mentioned also like how you he

22:40

asked his focus on tests. And

22:43

curious like. How do you

22:45

think about know? building an organization and

22:47

like company culture and in like, what

22:49

are sort of the main learning you've

22:51

had in that regard. So

22:54

far. Between. Bit cynics

22:56

and Tatar. We. Had

22:58

I'm really really really low. Number.

23:01

Of the the Saxons I think the

23:04

knob eileen in. In ten years we

23:06

could counted less than. Two.

23:09

Three percent. People. Speak

23:12

here. I

23:14

think the reason is the culture. Because.

23:16

Harrys really about openness.

23:19

Ah, I'm. Kind. Than kindness we

23:22

do is a first rule that

23:24

have so myself and the. And

23:27

a couple other people in them

23:29

in them chief positions, We.

23:31

Do every single finally interview.

23:34

right? So we still do all the fun

23:36

and the resealed the were growing. The

23:39

number one rule: Ease. Never

23:41

hire. Don't hire a thought

23:43

Six person. So.

23:46

You sometimes have amazing

23:48

talent. But. The behaviorally.

23:51

Could. Be challenging for the team. Or

23:54

you don't' If you've hired the wrong

23:56

person that a that person will affect. tenth

24:00

10 people around them. So

24:03

the ability, sometimes,

24:06

as I said, being a developer, sometimes you see

24:08

developers that get all cocky because

24:11

they are right and everyone else

24:13

is wrong or because they have

24:17

the only ones that

24:20

have the truth on how things

24:23

should be architected. So

24:26

we are really careful in hiring

24:28

people that understand the value of

24:31

collaboration within the teams. And

24:33

that's something that so far

24:35

over time allowed Bitfinnix to remain one

24:38

of the top platforms in the crypto

24:41

ecosystem while having probably

24:43

one tenth to one fifth

24:45

of the headcount

24:48

of other platforms. So

24:51

that's something about creating

24:54

a healthy team, healthy responsibilities

24:56

and chain of responsibilities. People are really

24:59

committed to the success of the company.

25:02

And so I think the

25:05

biggest takeaway from

25:07

my career is focus on the team,

25:09

focus on the people and you will

25:11

have success. I

25:14

really appreciate you saying that. Let's

25:16

talk about Tether.

25:19

How did Tether get started? Tether

25:21

started in 2014 as a

25:24

really simple product, a

25:26

simple idea. In 2014, there were

25:28

just a few exchanges, right? You

25:32

could count the exchange, the

25:34

cryptocurrency exchange, it was only on the

25:36

fingers of two hands. There

25:39

was Bitfinnix, of course, Bitstamp,

25:42

Kraken, Coinbase, there was

25:44

Okcoin and BTCC.

25:47

Not many others, at least

25:50

not many relevant others. And

25:52

so there was, at the

25:55

end of 2013, it was also the first moment in history that Bitcoin

26:00

broke $1,000 price. And

26:03

across multiple trading venues, right, so you

26:05

had maybe $1,200. At

26:08

any point in time, there was a

26:11

huge spread across exchanges, up

26:13

to 20%. Right. So

26:16

some exchanges were at $1,200, others were at

26:18

$1,950, others were at $1,000. The

26:23

reason is that by

26:26

then the work of the

26:28

arbitrageurs was not possible. The

26:30

arbitrageurs are those kind of

26:33

traders that sell Bitcoin

26:36

where on the exchange where the price is higher,

26:38

they send it for dollars, they take the dollars,

26:40

they move the dollars on the exchange where the

26:42

price is lower. They use these

26:45

dollars to buy Bitcoin and then send the Bitcoin

26:47

on the exchange where the price is higher and

26:49

so on and so forth. They do it in

26:51

the loop so that the

26:53

spread across these exchanges will go

26:55

down and they all get aligned.

26:59

In order to do this, you need

27:01

to be able to move Bitcoin

27:03

from one exchange to another, but

27:06

also you need to be able to move dollars

27:08

from one exchange to another. But

27:11

moving Bitcoin takes 10 minutes, right? The average

27:13

block down of Bitcoin is 10 minutes, but

27:16

moving dollars is extremely more difficult

27:18

sometimes. International wires might take one

27:21

day, two days, three days, seven

27:23

days, and there is the weekend.

27:27

Your trading opportunity, your arbitrageurs

27:29

opportunity is long gone if

27:31

you wait so much. The

27:34

really simple idea around Tether

27:36

was, okay, let's reuse

27:38

the incredible technology that

27:40

Bitcoin created called blockchain.

27:43

Let's just put the US

27:45

dollar because it's the most used currency in the world.

27:48

Simple as that. Back in

27:50

at that time, there was

27:53

no Ethereum yet. So The

27:55

way to do it, the only platform that

27:57

allowed that was called OmriLayer. That is a

27:59

color code. Point System. That

28:02

is based on bitcoin. So.

28:04

That are was born on Omni Lawyer.

28:07

Phosphor. Were many years and nearly. Petr

28:09

started as a. Symbol. Project

28:11

to allow our to make more

28:13

efficient trip the trading. But.

28:15

In two thousand and. Nineteen.

28:18

Two thousand and twenty We the and they

28:20

something changed. Your. With the.

28:22

Economy's. Old The words, especially in

28:24

emerging markets are developing countries. They.

28:28

They started to have big

28:30

issues were seen Argentina, Turkey,

28:33

Vietnam. When a swayed are

28:35

so many others in all these national

28:37

currencies are read, they were waiting against

28:39

the dollar really are adequately in some

28:41

cases, and I think that Argentina pays

28:43

us. Lost ninety percent against the U

28:45

S dollars five years. And. Are

28:47

more than eighty percent was lost by the Turkish

28:50

steer. So. All the

28:52

people who put people living those countries

28:54

where. Mean all these these people's

28:56

needs. Needed. Solution Needed a way

28:58

out. Needed. A way to

29:01

save. Their. Wells. Rise

29:03

of was not about speak racing series

29:05

about. A father of

29:07

a family that works for the entire year

29:09

to earn, for example, Turkish lira at the

29:12

end of the year is poor compared to

29:14

the beginning of the year. Just because it's

29:16

national, he's national currency when sound. The been.

29:19

As. So. People. Have.

29:22

A survival instinct And so they

29:24

started to look at socialists. As.

29:27

U S a T was already the most

29:29

you solution. That third that

29:31

crude the help them. So.

29:34

U S A T became basically desert

29:36

the sitting. With this currency that third

29:38

of ward the started to use. And.

29:41

to to build on all the in

29:43

old emerging markets now as if you

29:45

go in urgency and i'm wendell cyrus

29:47

you have find tanah places we can

29:49

you could even paying you a city

29:51

say our passing venezuelan so many other

29:53

countries and i must say we didn't

29:55

envision that we didn't think about it

29:57

when we treated tatar was read a

30:00

seems idea and so we are humbled

30:02

by the success of

30:04

USDT. I always say that before

30:07

2022 we never had even

30:10

a marketing team for Tether. And

30:13

so you know people started using

30:15

our product in the way they sold

30:17

more feet and in the way that

30:19

was more helpful to them. So that

30:21

is really exciting. Can

30:24

you give us an idea of how

30:26

it's used today? So kind of

30:28

in percentage, kind of like what percentage

30:30

is kind of like what

30:32

I would call normie use.

30:34

So kind of like the

30:36

Turkish father or the Argentinian

30:38

family mother who kind of

30:40

keeps their salary in Tether

30:42

in order to spend it as efficiently

30:45

a couple of months later. So

30:48

it is really just to

30:50

say it's really hard to pinpoint

30:52

exact figures because we

30:54

are leaving another centralized world and

30:56

so information is captured. But

31:00

as of today I think

31:02

that at least 50% of

31:04

the Tether in circulation, the

31:06

USDT in circulations, in

31:08

circulation are kept as

31:11

savings and not necessarily

31:13

used for trading. So that is a

31:16

great achievement because I think

31:19

it happened in four years. In

31:21

four years people had this incredible

31:25

bad situation of financial

31:27

suffering and I

31:29

always say that we

31:32

are kind of in a weird situation

31:35

at Tether because we see

31:37

that if we have more success means

31:40

that the world is going

31:42

to go towards the worst place. Because

31:46

that's the reason for all

31:48

these countries to look at the dollar is because

31:50

something is not working in their own country and

31:52

the easiest way to have access

31:55

to the dollar is USDT. So

31:57

in one way someone could see it of course is a

32:00

success for a company but

32:02

it's hard to understand that

32:04

success is also determined by

32:08

imparting competency from

32:13

financial management of many countries

32:15

and the inability to preserve

32:18

wealth from

32:22

government in these countries that led

32:25

people to fly towards a safer

32:27

currency. You've moved stack

32:29

or you have kind of added stacks

32:32

over the last couple of years. So

32:34

as you said, kind of you guys started out on Omni

32:37

and then you added Ethereum and now they see

32:40

USCT is natively issued on a

32:42

number of networks. How

32:44

do you determine which chains to

32:46

issue natively on?

32:48

So basically if you look at

32:51

payments on chain, Tether on Tron

32:53

is kind of where it's at

32:55

the moment. So how

32:57

do you kind of see these movements? How

33:00

do they happen? Do you plan

33:02

them? Yeah. So

33:04

we listen to

33:07

communities and every

33:09

time we deploy USCT on the chain,

33:11

we do an enormous amount of

33:13

the diligence that goes

33:15

from the security of the blockchain.

33:17

So how is the technology

33:21

built to the interest

33:23

of a real community and

33:26

the actual ecosystem around such

33:29

blockchain? It's pointless

33:31

to list, to add USCT to a

33:33

chain that doesn't have users

33:35

or interest because then we expose

33:39

ourselves to risks while not

33:42

getting any benefit in terms of adoption. So

33:45

the interesting and I get this question a lot, the

33:47

interesting thing about Tron is that we

33:50

launched first Ethereum. Well, first there was

33:53

Tether Omni, then there was

33:55

USCT was launched on Ethereum

33:57

and then was launched on Tron. People

34:01

choose whatever they want, right? So people

34:03

choose the transport layer of USD they

34:05

want. We, just

34:08

to be clear, we are centralized stablecoins

34:10

so all our features

34:12

and capabilities are shared across the

34:14

different blockchains. For example, we can

34:16

freeze us. It's public,

34:19

right? We can freeze wallets

34:21

working with law enforcement. That

34:24

is something that applies to all the different blockchains.

34:26

And so people just choose the

34:29

transport layer, so the blockchain for

34:31

USDT, that is more fitting to them.

34:34

And 99% of the people would

34:37

always choose something that is faster and something

34:40

that is cheaper. Let's

34:42

think about how, what I said before.

34:45

So Tether USDT has becoming

34:48

really important for hundreds of millions of

34:50

people across the board, especially in emerging

34:52

markets and developing countries. We

34:55

don't support US customers and

34:58

I don't think Europeans need

35:00

stablecoins either. That is my

35:04

take. So the

35:06

vast majority of Tether users is

35:08

all in emerging markets and developing

35:11

countries. Countries that, as we said,

35:13

their national currencies is losing

35:15

really quickly against the dollar and

35:18

where every

35:20

dollar or transaction fee is

35:22

extremely important. We have

35:24

seen after 2017 how many times

35:26

the theorem gas fees, they grew from

35:31

$2 to $50 to $100. Every time there is some excitement,

35:39

the gas fees go crazy high. And

35:42

so Tron had really

35:44

low cash fees for a

35:47

long time. Tron had

35:49

these more centralized as these system of

35:51

validators and for a long time had

35:53

proof of stake compared to proof of

35:55

work. The theorem moved from proof of

35:57

work to proof of stake more recently.

36:00

So, you know, the throne

36:02

was a simplified approach to EVM

36:04

that was extremely effectful because, you

36:07

know, all these in think about

36:10

Africa and South America and the

36:12

Central America, certain countries

36:14

of Asia, you know, even 50 cents of transaction

36:17

fees are really high, right? So

36:19

maybe they earn $50 to $200 in a month.

36:24

And if they have to spend $5 every

36:26

time they send a transaction, that

36:28

is not a product for them. The

36:31

interesting and sometimes, you know, the Ethereum community

36:34

criticized us a

36:36

little bit because, you know, the throne

36:38

is still heavily used. The

36:40

reality is that Ethereum

36:43

had two, four, Ethereum took four

36:45

years to come up with layer

36:47

two solutions that would

36:49

bring down the fees to

36:51

a variable value, right, for the

36:54

vast majority of the users of blockchain.

36:57

And so for if you give

36:59

to a competitor that will be trying this

37:01

case four years of first mover advantage, of

37:04

course, it's really hard to take it

37:06

back, especially when also all these layers

37:08

to solutions built on Ethereum are completely

37:11

are anyway competing

37:13

with each other as well. It's like, you know, the

37:15

Simpson meme where, you know,

37:17

with the Scots where basically in the

37:20

end everyone hates each other.

37:22

And the reality is

37:24

that we are

37:26

seeing, you know, we think that it is

37:30

really important for us to

37:32

be more diversified on different

37:34

blockchains. It is

37:36

important that although these blockchains will start

37:39

to grow and ecosystem will keep transaction

37:41

fees low and will

37:43

keep the transaction speed high in

37:45

order to fulfill the interest and

37:47

the need of

37:50

emerging markets populations. TSE20

38:00

transfer current is around $1.50. So

38:02

it should be

38:04

impacting the people that

38:07

are served best by Tether. Yes,

38:09

exactly. And that's why we are seeing

38:12

opportunities on launching

38:15

another change recently. We launched on

38:17

Celo. Well, two days

38:19

ago, we launched on Celo that has lower transaction

38:21

fees, is VVM compatible. And we

38:23

are also looking at Polygon and

38:25

other layer tools. What I'm saying

38:27

is that now

38:30

that hundreds, well, actually tens

38:32

of thousands of integrators like

38:34

merchants, payment solutions have adopted

38:37

Tron is really hard for the layer

38:40

two solutions on Ethereum to compete. They

38:43

had four years of first

38:45

mover advantage. So

38:49

you started focusing

38:52

on this arbitrage use case, and

38:54

now there is, you know, kind

38:56

of Tether is like branched out and

38:58

there's a lot of this, you know, payment use cases

39:00

or people using it as a store of value. I

39:03

heard in some interview,

39:05

you talked about

39:07

Tether as a

39:09

sort of over collateralized bank

39:11

account. I'm curious, if you

39:14

think of like the long term vision

39:16

for Tether, like, where do you see

39:18

it going? So

39:22

Tether evolved a lot. I mean, we

39:27

were the black sheep of crypto.

39:29

Maybe we still are and I'm

39:32

not realizing it, but we

39:34

were the black sheep

39:36

of crypto for a long time. You know, I

39:39

think the mistake we made is

39:42

that we thought that

39:44

keeping your head down and, you know,

39:47

not being, you know, not

39:49

entering in public disputes and

39:51

not being, you know, present

39:54

on Twitter and so on would

39:56

work for you.

40:00

for us, you know, it's like,

40:02

okay, if I work, if I keep my

40:04

head down and do a good job, people

40:06

will be okay with me, we'll be happy.

40:09

And we realize that we are good people.

40:11

And then, you know, that didn't work

40:14

well. And, you know, we started to understand

40:16

that people needed heavy

40:19

quantities of transparency and I think,

40:21

you know, it's, it's rifled to

40:24

ask for that. And so

40:27

we changed, I think, heavily in the

40:30

last years, the amount of information that we

40:32

publish, the way we managed

40:34

our services much more public, we

40:37

came up, we were the first ones to come

40:40

up with an attestation with a breakdown of the

40:42

reserves showing how much we have in different things.

40:45

We were subject to many critics that

40:47

we took well and we applied to

40:49

the company so that we could improve

40:52

also our service, you know, I'm sure you

40:54

remember the commercial paper part, you

40:56

know, although there were a lot of

40:58

nonsense like we had ever grande and

41:00

you know, all the Chinese papers and

41:02

so on that was very stupid. But

41:04

anyway, we we

41:06

proved that we could convert everything

41:09

almost everything in US TBLs. As

41:11

of today, so the community was asking

41:14

us to do changes. And we always listened

41:16

to the community we made these changes. As

41:19

of today, I think we have well as

41:21

the 31st of December,

41:23

2023, we had $80.3 billion

41:28

US Treasury bills that would put us

41:30

at the 20th as the 20th

41:33

country in the world as owners

41:35

of a US debt. And

41:38

when it comes, I think to the three

41:41

months Treasury bills

41:44

owners, we are the third one, as

41:46

in terms of holdings. So

41:49

at our at the majority of our

41:51

TBLs are all short terms like within 90 days.

41:54

And so now nowadays, in

41:57

the last attestation, we published other

42:00

numbers like we had 5.2 billion on

42:02

top of the 100% reserves that are covering all the

42:09

issued USDT tokens. So

42:12

meaning that we had,

42:15

Tetris is making good money, right? In a

42:17

quarter we made almost $3 billion. And

42:20

we could have dividend any other company, any

42:22

other normal company, imagine a bank that would

42:24

make $3 billion in profits. And oh my

42:26

god, they would distribute these profits

42:29

left and right to all the shareholders up

42:31

to the last cent. But

42:33

no, with Tetris we decided to keep the vast

42:36

majority of these profits within the company up to

42:38

5.2 billion as the last

42:40

quarter. So that

42:42

ran Tetris end up

42:45

in being 5.2 billion over

42:47

collateralized. So of

42:50

course banks are upset because banks

42:52

are lending out 90% of their

42:55

portfolio and their balance sheet. And

42:58

Tetris that is doing the opposite is

43:00

actually accruing more money and keeping it

43:02

to show that you can

43:04

build something that is safe. You can build a

43:06

financial tool that is

43:08

safe and doesn't

43:10

need to lend out people's money.

43:13

And so

43:15

that's what I like

43:18

about Tetris, right? The

43:20

idea of revolutionizing

43:22

how some

43:24

certain things in finance are not. So

43:28

the yield on US

43:30

treasury bills has gone up enormously

43:33

over the last year or so. Have

43:35

you ever thought about kind of making

43:37

USCT in some form yield bearing

43:40

and letting people who actually use

43:42

it and hold it participate

43:44

in that windfall? So

43:48

that is a really interesting one, right? So

43:50

I get this a lot as well. The

43:53

reason why, there are two different reasons why

43:55

we cannot do it and doesn't

43:57

make sense for us to do it. The

44:00

first one is that if you start

44:03

giving out interest and

44:05

sharing interest, that becomes a financial

44:07

product, that then becomes a

44:09

security. So that's

44:11

something that is, I don't think, is

44:13

a good approach. So

44:16

I think that many of the products that

44:19

are the stablecoins that are trying

44:21

to do a stablecoin that

44:23

provides an interest are

44:25

going to end up in having some

44:27

issues with the

44:29

US regulators and other regulators. But

44:32

also, I consider

44:35

myself a scientific person. And

44:38

when I create a product, I always think

44:41

how for whom this

44:44

product is created for and who

44:46

is going to use this product.

44:50

If you live in Argentina, where the

44:52

intra-derivability of the Argentina basis is higher

44:55

than 4% to 5%, does

44:58

this really matter that at the end of the year you get 4%? And

45:01

on the other side, if we get that 4%, we

45:04

can create a really incredible

45:06

stable product that is

45:08

helping hundreds of millions of people. So

45:12

that's why I think if

45:14

you are in the US, you

45:16

have already the best banking rails. You

45:18

have already the US dollar. So if

45:21

you put money in a stablecoin,

45:23

you expect to receive an interest,

45:27

because that's the concept of a saving account and a checking account. On

45:30

a saving account, you want to have an interest. But

45:33

everywhere else outside the US,

45:37

for all the people outside the US, the interest

45:39

is in especially the ones living

45:41

in the merchant markets and developing countries that

45:43

are the actual users of a

45:46

stablecoin. Then that

45:48

4% is not really meaningful,

45:50

because what they care about is to

45:52

have something that protects themselves from a

45:55

much worse situation. So

45:58

I totally understand. And that

46:00

reasoning, of course, one

46:03

thing that did happen like a few years ago,

46:05

that sort of like

46:07

along these lines, right, is that a

46:09

lot of the usage of

46:11

Tether or of USDT was that people hold

46:14

it on exchanges, right, the trade. And

46:17

then I think, you know, Binance created

46:20

their BUSD. And

46:22

you know, I think it was because

46:24

they say like, well, all these people

46:26

holding, you know, USDT there and of

46:28

course, the sort of profits of that,

46:31

you're good with Tether. And they

46:33

were like, okay, let's if we can swap this

46:35

out with our own stablecoin, right, we can we

46:37

can capture that. So I'm curious,

46:39

do you think that might be a path

46:41

for basically USDT would incentivize

46:43

like some of these large

46:46

platforms where massive amounts of

46:49

USDT get held or do you see that as

46:51

a sort of competitive threat? Well,

46:54

I know that some of our our

46:57

biggest competitors are kind of doing that, right.

46:59

So, you know, I know that from

47:02

that, some of our competitors are

47:04

paying big institutions and

47:07

exchanges to hold their

47:09

stablecoin. Again, to

47:13

me and to, you know,

47:15

many legal teams that we

47:17

talk to, that ends

47:19

up in you are giving interest

47:21

to someone else, hence can

47:24

become or can be considered as equity.

47:26

So the point is that

47:30

Tether today is one hundred two billion

47:32

dollars in market cap. Maybe

47:34

we'll go down, maybe someone else will create a

47:36

better solution. But we are not. Look,

47:39

again, we are in for the tech,

47:42

we are keen for the innovation. If we

47:44

become the second biggest stablecoin or third biggest

47:46

stablecoin, as long as it's saying as long

47:48

as it follows our ethos, that's

47:50

fine. We are happy with it. Right. So

47:53

we don't we don't have to be forever the

47:56

first. We need what

47:58

we like is to. leave a sign

48:00

in terms of innovation, a utility,

48:03

and how with our technology,

48:06

with our passion, we can change the

48:08

world. And if someone will create

48:10

something better, that's fine. But

48:12

I believe that creating, trying to compete

48:18

and trying to reduce

48:21

the security, also in this case

48:23

from the legal side, right? So

48:26

if you start doing more and more things, if

48:28

you keep adding flavors to

48:31

the stable coin, and you start doing this

48:33

and that because you are scared to become

48:35

the second or the third or losing market

48:37

cap or losing market share, look,

48:39

I mean, that's not

48:42

us, right? So we believe that

48:44

the simpler it is, the

48:46

safer it is. So

48:48

if you start adding all these nuances,

48:51

eventually you will fall in an issue,

48:53

and that could endanger the stable coin.

48:55

So again, Tether has more

48:57

than 300 million users across the world. And

49:00

so the only thing we do care about is that is safe.

49:05

Your main USD saver

49:07

coin competitors, USDC, how

49:10

do you see the differences between USCT

49:12

and USDC? Well,

49:16

they are focusing mostly, and historically

49:19

they have been focusing a lot on the US.

49:22

To me, and Europe

49:24

now, to me is

49:27

if you look at the US, and if you look at

49:29

Europe or Switzerland, everyone has

49:32

two credit cards in or debit cards in their

49:34

pocket, have cash, have two

49:36

bank accounts, you know,

49:38

there is no need. To me,

49:40

I always described like trying to sell an

49:42

ice cream to an Eskimiz, right?

49:44

So there is no point. When

49:47

you create a product again, you have to create

49:49

it for the people that really

49:51

need it. You know, and

49:54

those are the ones that don't

49:56

have access to a bank account. There

49:58

are billions of people in the world. they don't have access

50:00

to a bank account. There are billions

50:03

of people that are not bad

50:06

people, they're really good people, they're nice people,

50:08

but they are not

50:11

really interesting for the banking system. Because

50:13

for the banking system, if you're not

50:15

profitable enough as a human being,

50:18

you will not be on board. If

50:20

you're not putting all your salary and if your salary

50:23

is not enough, is not big enough to

50:25

justify their outdated

50:28

financial infrastructure, then

50:32

they will never accept you as a customer. For

50:34

all those people, there is USD. We were

50:36

clear when we saw the pandemic

50:39

and all these emerging market issues happening,

50:41

we were clear. That is our crowd,

50:43

that is the people we want to

50:45

serve. We see

50:47

our competitors always trying to play

50:51

the game of Wall Street and we

50:53

want to be the stablecoin of institutions.

50:55

But these institutions have

50:57

already the best banking rails. So

50:59

the biggest difference in strategy is we

51:02

want to be the last stablecoin, the

51:04

dollar for the last mile. For

51:07

the normal people, we think that bankers have already

51:10

too much. What

51:17

are your thoughts on decentralized stables?

51:21

So basically things like

51:23

Dye, Ry, I mean,

51:25

so basically there's different flavors, right?

51:27

Like fully collateralized, collateralized with assets

51:30

that may be difficult to get

51:32

a price feed for under collateralized,

51:34

algorithmic and so on. Do you

51:36

see those gaining ground in any

51:38

way? Because kind of the way

51:40

that centralized stablecoins work

51:43

is we leverage heavily

51:45

the existing financial infrastructure

51:47

and kind of are

51:49

exposed to financial infrastructure

51:51

risk that kind of pertains to

51:53

the legacy system. How

51:56

do you see that play out in the future? Let's

51:59

go with them. a little bit of history, right?

52:01

So while we think about

52:03

algorithm-stable coins, we think about TerraLuna. We

52:06

all know how it ended up, right? So

52:09

I was quite vocal

52:11

in the months before saying

52:14

to many people, look, TerraLuna

52:16

is going to blow up.

52:19

And many answer me, well, you are

52:22

going to say that because TerraLuna

52:24

is going to eat up all

52:26

Tether market share. But

52:30

the interesting part is that it's

52:33

easy to create a stable coin that is, you know, 1

52:36

billion in market cap, 2 billion, 3 billion,

52:38

5 billion. If

52:40

you get to 10 billion, or I think 17

52:43

to 18 billion was TerraLuna

52:45

at the peak, the

52:49

liquidity has to be there, right? You have to

52:51

be able to pay out immediately.

52:53

You have to be able to redeem.

52:56

In 2022 was April-May when TerraLuna blew

53:00

up. There was, you

53:02

could see the difference between an

53:04

algorithm-stable coin that was

53:06

made through

53:10

some, you know, weird incentivization

53:14

mechanisms and Tether.

53:17

So TerraLuna fell

53:20

because they couldn't sustain

53:22

hybrid entries. So

53:25

everything started to blow up.

53:28

After TerraLuna blew up, publicly,

53:32

a group of hedge funds started

53:34

to short USDT

53:37

on the secondary market. So

53:39

you could see, and

53:42

recently, I think DCG came

53:44

out or neither were some

53:47

disclosures during a class action showing

53:49

that also DCG shorted,

53:52

during those times USDT for 400

53:54

million dollars. But

53:58

there are other hedge funds like fear3 and and

54:00

others, they shorted billions of

54:02

dollars all together across all these groups.

54:05

In a few days of USDT, they

54:07

kept the price of USDT below one

54:09

dollar on the secondary markets. Why they

54:12

did that? Of

54:14

course, they did that because they wanted to

54:16

close a bank run. That

54:18

is what basically happened on

54:21

Terra Luna. So they wanted to

54:23

close a bank run where people, if you

54:25

keep the price of USDT below the

54:27

dollar or a stablecoin

54:29

below the dollar, what happens is

54:31

that market makers are coming in

54:34

on the secondary markets and buying

54:36

the stablecoin for cheap, going

54:38

to the stablecoin issuer, i.e., in this case,

54:40

Teller, redeeming for $1,

54:43

getting the dollar, moving the dollar back

54:45

on the exchange, buying cheap stablecoin, going

54:47

to the stablecoin issuer, redeeming it, and

54:49

so on and so forth. That

54:51

goes, that is, in the

54:53

banking world, the start of

54:55

a bank run. So what happened

54:57

with Teller in 2022 in

55:00

May was that Teller was able to pay out

55:02

in 48 hours, $7 billion. That was 10% of

55:07

our reserves. And in 20 days, we

55:09

paid around $20 to $25 billion of

55:12

redemptions. So that was 25% of

55:15

our market cap. So

55:17

there are a few banks in the history

55:20

that have been subject to a bank run. Washington-Muthwell,

55:23

2008, 10% redemption, they

55:25

went belly up. Silicon

55:27

Valley Bank, Sinsere, Silvergate, all

55:29

these guys didn't survive

55:31

to bank runs. We

55:34

are seeing recently also another

55:36

bank in these days had issues.

55:38

So the

55:40

difference between a

55:43

stablecoin that is backed

55:45

by liquid and real

55:47

world assets and an algorithm-stablecoin

55:49

is enormous. I don't think an

55:52

algorithm-stablecoin makes sense. A war can

55:54

grow above a certain threshold

55:56

because otherwise become too risky. attackable

56:01

by attackers. They tried with us, they

56:03

couldn't make it and

56:05

they lost enormous amount of money

56:07

because we were backed and we

56:09

were solid. They

56:11

had and when

56:14

they tried to attack Terra Luna they would

56:16

succeeded. It blew up really really fast. So

56:19

when it comes to the other times like collateralized,

56:22

like DAI, I think they are

56:24

really interesting concepts. The issue is

56:26

how you grow the market cap,

56:29

having extremity volatile assets in your

56:31

portfolio. Like if you have

56:36

Bitcoin and Ethereum as part of your collateral

56:38

that becomes hard because we have seen how

56:40

in the last four years Bitcoin

56:42

and Ethereum lost 80% of their

56:45

value. So it's really really

56:47

tricky. So that pushed DAI into

56:49

my understanding to move towards TBLs

56:51

as well. But if

56:53

you start

56:55

to using TBLs as part of your collateral you

56:58

are no different than Tether or the other centralized

57:01

stablecoins. So I think in a

57:03

way DAI

57:06

decided that the model of Tether

57:08

was probably the right one and

57:10

more scalable one in the long

57:12

term. Yeah, they added GSDC

57:14

as collateral fast. So

57:16

the T-Buds I think this is something

57:18

that's kind of slowly ramping up but

57:20

for some time 70% of DAI collateral

57:22

was actually USDC. And

57:28

so it's just a proxy right? So I think look

57:32

the point is and the

57:34

question that people should ask themselves is why

57:37

you need a stablecoin? So

57:40

what you want to get from a stablecoin? Because

57:44

if you want to have an asset that

57:47

is resistant to the wrath

57:49

of God that no one can take away

57:51

from you and so on. That is Bitcoin.

57:53

So you have already something that is digital

57:56

that is unconfiscatable

57:58

that is built to

58:02

survive to all

58:04

the crazy monetary policies that WORD

58:06

is going to. If

58:10

you want a stablecoin that quacks like a

58:12

dollar, probably you are still bound

58:15

to the fiat WORD.

58:17

So you need for a specific use

58:19

case, a dollar. So

58:23

why we are trying to push

58:26

all these crazy mechanisms to

58:28

try to recreate the world, recreate a dollar and

58:31

not using, you know, if you want a dollar, if

58:33

you're trying to recreate the dollar, it means you need

58:36

the dollar for something and you need

58:38

at some point to exchange that dollar for a

58:40

dollar, right, for a real world dollar. So

58:43

if you are in that situation, probably

58:45

you just want something that has, that

58:48

is backed by a

58:50

dollar or backed by the

58:52

closest proxy to a dollar that is USD

58:54

bills. So you

58:57

know, you why putting Bitcoin in

58:59

Ethereum to recreate the dollar? I

59:03

mean, when eventually that dollar that you created through

59:05

Bitcoin in Ethereum, you have to sell it for

59:07

a dollar that is backed by the real dollars

59:10

or by the US economy that would be tethered,

59:12

right? You cannot cash out easily die. You have

59:15

to sell it for USD. You cannot redeem it

59:17

directly for USD. And so that's

59:19

the point of the stablecoin is to be able

59:21

to have a link to

59:24

the physical world or to the fiat world. So

59:27

that's that's my small rant around

59:29

the stablecoins. It's just not that we didn't think

59:32

about how we can create something

59:34

better. Well, here are some ideas that are

59:36

quite interesting. But the reality

59:38

is that nothing is really proven to

59:40

be able to become so big as

59:42

USDT and maintaining

59:44

stability no matter what that

59:47

without using a heavy amount

59:49

of physical assets. So

59:54

US dollars, I mean, especially in

59:57

this crowd, they are

59:59

heavily criticized. They

1:00:02

are often made out to be something that could

1:00:04

collapse any day now. So

1:00:07

have you thought about adding

1:00:09

an additional product that doesn't

1:00:11

just use US dollars or

1:00:14

treasury bills as the

1:00:16

underlying collateral by having some sort

1:00:18

of basket of currencies or special

1:00:21

drawing rights and basically

1:00:24

adding the Swiss franc and the euro

1:00:26

and the yen and so on to

1:00:28

the mix? So

1:00:30

this is a great question and there is

1:00:32

a lot to unpack. The thing is that,

1:00:35

so again, if you are scared

1:00:38

about the US dollar, there is a

1:00:40

way you should be

1:00:42

scared is that eventually, like in the

1:00:45

Weimar Republic, you might be able to

1:00:47

buy a loaf of bread with one

1:00:49

billion dollars, right? But I

1:00:52

can tell you and if the

1:00:54

dollar gets to that point, the

1:00:57

euro is long gone. And

1:00:59

that's not about speak about the Japanese

1:01:01

yen. Well, the Swiss franc

1:01:03

is probably the only thing that will remain

1:01:06

and will survive because Swiss people understand finance

1:01:08

better than anyone else. But

1:01:11

there is no other basket. The only thing that would

1:01:13

make sense is gold. What

1:01:17

about a basket of kind of

1:01:19

like stocks and, you know, in

1:01:21

principle, I mean, you would ask

1:01:23

yourself, where would the value go?

1:01:25

Right. And basically, physical values would

1:01:28

still be there. So kind of wouldn't

1:01:30

you want like shares in

1:01:33

companies and actual

1:01:35

physical goods like Goer,

1:01:37

Silver, and I mean, there are

1:01:40

representations of those on chain

1:01:42

as well, right? Well,

1:01:45

if the dollar goes bust, that

1:01:47

is the entire New

1:01:50

York Stock Exchange and all this, the

1:01:52

American markets are the biggest ones. So

1:01:54

do you want ready to have some

1:01:57

stocks in the US market? I

1:02:00

think if we are thinking, and I like the

1:02:03

thought, right? So about what will happen

1:02:05

if something happens to the US dollar.

1:02:08

But that means that the entire world

1:02:10

economy is going towards a reset.

1:02:13

And there are only two things that will save you, Bitcoin

1:02:16

and gold. So every

1:02:19

human, humanity went through many

1:02:21

resets and they will, they

1:02:23

always were done through gold, never

1:02:26

through fiat currency. Now

1:02:28

we have also Bitcoin as a great

1:02:30

alternative and gold on steroids.

1:02:34

But share the thing, right? So USDT

1:02:36

has to represent a dollar. So

1:02:39

with one USDT, you shouldn't be able to buy more

1:02:42

than one dollar, right? So for us, the

1:02:44

USDT is a representation of dollar that

1:02:46

is a representation of the US economy.

1:02:49

If again, with the dollar you can

1:02:51

buy, if you need $1 billion to buy

1:02:53

a loaf of bread, you will

1:02:56

need 1 billion USDT to buy a loaf

1:02:58

of bread. It's not our business to make

1:03:00

the dollar prettier and more

1:03:02

solid than the dollar itself, right? So it's

1:03:04

just a representation of dollar. If

1:03:06

you really want something that is better, then

1:03:09

you might want to use Bitcoin or

1:03:12

gold. Yeah, I

1:03:14

mean, I think a huge part of the

1:03:16

value of something

1:03:18

like USDT is also, well, I mean,

1:03:22

let's say in crypto, USDT or

1:03:24

USCC is used a lot, right? Like let's

1:03:26

say you invest in some company or you

1:03:28

do something like that. It's mostly done with

1:03:30

stablecoins. And of course, the nice thing

1:03:32

is you will have a contract and in a contract

1:03:34

it says, oh, $50,000, $70,000, something like that. And

1:03:38

you can just send $70,000 USCT and

1:03:41

there's no complexity, there's no conversion,

1:03:43

accounting is simple. So

1:03:45

like I think even if you created

1:03:47

some kind of other stablecoin that, you

1:03:49

know, let's say tracked inflation and it's

1:03:53

from a practical perspective, it

1:03:55

will be less useful, right? Then something that

1:03:58

just mirrors the dollar. I

1:04:00

couldn't agree. The point, the question that people

1:04:02

should always ask themselves is what

1:04:04

I'm using this for. So if

1:04:07

you're using it to just to have

1:04:10

a dollar, then you should go for the closest

1:04:12

thing to a dollar that you can find. And

1:04:14

that will quack as a dollar and also

1:04:16

in the accounting, as you said. So

1:04:19

for the rest, there is plenty of other options

1:04:22

that that we created in the last many years.

1:04:25

Cool. Well, let's let's move on to

1:04:27

the last topic because I mean, we've

1:04:29

talked about Bitfinex, we've talked about tether.

1:04:31

So I mean, you already are doing

1:04:33

a lot. But then I think you

1:04:35

started another thing which is called

1:04:38

hole punch, which is

1:04:40

focused on building peer

1:04:43

to peer applications. Can you talk

1:04:45

a little bit about what is hole punch and,

1:04:47

you know, the kind of related things to hole

1:04:49

punch and like what's the vision behind that? Sure.

1:04:53

So with tether, we

1:04:56

came from this enormous learning

1:04:58

curve on how

1:05:01

the world is going bust. And

1:05:03

it's not not just us, right? So then one

1:05:06

of the main reasons of crypto is that we

1:05:08

think that this many of us think that economy

1:05:12

is the real world economy

1:05:14

is going towards the direction where, you know,

1:05:18

countries are more and more

1:05:20

at war. Among

1:05:22

themselves, the economy

1:05:24

is getting more unpredictable. And

1:05:26

so we created a solution

1:05:29

or to that that is, you know,

1:05:31

Bitcoin or blockchain based products and so

1:05:33

on. But, you

1:05:35

know, we have this concept in crypto

1:05:37

that is called individual sovereignty, where

1:05:40

you should be, of course, we all live

1:05:42

in a society, but you should

1:05:44

have financial freedom. So you need

1:05:46

to you want to have the direct to

1:05:49

interact with whoever you want.

1:05:51

But the individual sovereignty

1:05:54

that through financial freedom is

1:05:56

incomplete, you need also freedom of speech

1:05:58

to achieve that. Because

1:06:00

if you have individual sovereignty,

1:06:02

but you cannot say anything to

1:06:05

anyone because otherwise you get censored

1:06:07

or going jail, that

1:06:09

doesn't matter much, right? And

1:06:11

vice versa if you have freedom of

1:06:13

speech, but then you don't

1:06:15

have financial freedom, whatever

1:06:17

you say can be used to seize your

1:06:20

funds or make sure that you are

1:06:22

not really free to use your own money. So,

1:06:25

five, six years ago, myself

1:06:28

and Matthias Boos, the developers,

1:06:31

both really enjoying the

1:06:34

concept of peer-to-peer communications as we

1:06:36

are enjoying the concept of Bitcoin

1:06:38

as peer-to-peer money. We

1:06:42

both came through from a background of five

1:06:45

sharing system like BitTorrent, you know,

1:06:47

and all the predecessors. And

1:06:50

we know there is an interesting aspect of

1:06:53

five sharing systems that I think is really

1:06:56

was comparable and is

1:06:58

comparable to blockchains

1:07:01

and the money

1:07:03

systems. So you

1:07:05

might be familiar with Napster, the first

1:07:08

five-string system, you know, went bust, closed,

1:07:10

and all through different iterations, developers

1:07:13

tried to create different solutions to

1:07:15

make five sharing more

1:07:17

decentralized, exactly as we

1:07:20

tried to do with money. So,

1:07:23

different attempts, right, from you

1:07:26

had Linewire, you have Kazam,

1:07:28

you have a donkey, you

1:07:30

have a mule, and then finally

1:07:32

you had BitTorrent. The

1:07:34

difference across all these different approaches

1:07:38

is that all the different approaches before

1:07:40

BitTorrent, they had a centralization

1:07:42

point. They had, you know,

1:07:45

if you were using Kazam

1:07:47

or Linewire, you wanted to look

1:07:49

for a movie, you had

1:07:51

to connect to a

1:07:54

centralized index, like, or certain servers that had

1:07:56

all the list of all the files, you

1:07:58

can find, download the files, right? And

1:08:01

all these centralized servers were shut down. So

1:08:04

then people jumped to the

1:08:06

next version, trying to decentralize file sharing a

1:08:08

bit more, and so on and so

1:08:10

forth. There was this history of

1:08:13

file sharing. I arrived to the

1:08:15

point where with BitTorrent, everything was

1:08:17

really decentralized from the indexes. So the

1:08:19

list of all the files were

1:08:21

decentralized through a system called distributed

1:08:24

hash table, DHT. The

1:08:27

file sharing itself, the exchange of the files

1:08:29

was completely peer to peer, and

1:08:32

then so on and so forth. Now, with

1:08:34

Matias myself, really

1:08:36

were fun of BitTorrent. And so we thought,

1:08:38

what if we took

1:08:40

the very same technology around BitTorrent, but

1:08:44

we improve it, we make it

1:08:46

available for not only for file sharing,

1:08:48

but for real time streams. In

1:08:51

the world, almost everything is real time

1:08:53

streams. This chat on the

1:08:55

river side is a real time stream, right? So we have

1:08:57

tens of streams, audio, video, whether

1:09:00

we browse this, the internet is a

1:09:02

real time stream. And so we thought

1:09:04

if we can take those protocols and we can

1:09:06

create a new set of protocols that are free,

1:09:09

that are open source, that are

1:09:11

private. So we took also all the

1:09:13

learnings from blockchain to create identity,

1:09:16

to use that encryption that we learn

1:09:18

from blockchain to make it more secure

1:09:21

than BitTorrent was and is,

1:09:23

right? So we spent

1:09:26

five years to recreate the basic

1:09:28

protocol technology to create

1:09:31

a better layer for internet. I

1:09:35

think personally, when the technology is,

1:09:38

and I think that something

1:09:41

that really annoys me is

1:09:43

the fact that internet changed

1:09:45

from the beginning from the real purpose

1:09:47

it was born for. When internet was

1:09:49

born, was born to allow people

1:09:52

to talk to each other, share information with

1:09:54

each other. Every computer had

1:09:56

an IP address and still today has an

1:09:59

IP address. so that every

1:10:01

computer could directly interact with the

1:10:03

others. But more and more after the

1:10:05

year 2000, centralized

1:10:08

systems were born, like, so, Google,

1:10:11

of course, then email,

1:10:13

super centralized, and then, you

1:10:16

know, WhatsApp, Telegram, all these

1:10:18

communication systems are incredibly centralized.

1:10:20

They are going actually against, and all the

1:10:22

cloud providers are incredibly centralized. Now, 70% of

1:10:24

the entire traffic intern is

1:10:28

provided by servers that are

1:10:30

running on the top three

1:10:33

cloud providers, AWS, Microsoft,

1:10:36

Google. So is

1:10:38

that really the future

1:10:40

that we want? So what

1:10:42

happens is, and now, we

1:10:44

go back to the beginning of this, this, this chart.

1:10:47

We are building internet for the best

1:10:50

case scenario where everyone is friendly to

1:10:52

each other. What if

1:10:54

tomorrow, and we should learn from history,

1:10:56

what if tomorrow, Italy

1:10:58

and France will not be friends with each other,

1:11:00

or, I don't know, Germany

1:11:03

and Spain, or, I don't know, the

1:11:05

US and another country? So

1:11:07

what if, and we build,

1:11:10

and, you know, the reality is that all

1:11:13

these solutions, like WhatsApp, are

1:11:15

built by a specific company, a

1:11:17

single company, and is using

1:11:19

data centers in specific locations, in

1:11:22

specific geographic locations. So you

1:11:24

might see a future where data

1:11:26

will be held hostage and

1:11:29

used against other countries. If

1:11:32

you live in Europe, you use WhatsApp. If

1:11:34

you live in Rome, you know,

1:11:37

use WhatsApp. Every

1:11:39

single time you send a small message, it

1:11:41

will go to Frankfurt and go back to

1:11:43

Rome, right? So 90% of the people, for

1:11:45

90% of their time, are

1:11:48

talking to people that are within, you know, two

1:11:50

kilometers from where they live. Yet,

1:11:53

every time we chat, data

1:11:56

travels thousands and thousands of miles,

1:11:58

just to go back. you

1:12:00

know, two-hour neighbor or to go back to

1:12:02

be two kilometers per month. Is

1:12:04

that intelligent? No, of course. Imagine

1:12:07

how much governments are spending in

1:12:10

internet infrastructure just to create

1:12:12

beefier lines, bigger internet lines

1:12:14

for people that are talking for most

1:12:16

of the time with their neighbors or with their families

1:12:19

that are living in the same area. That

1:12:21

is really utterly stupid, sorry to say that, but

1:12:24

it is. And

1:12:26

so this is done because

1:12:28

centralized companies and big tech

1:12:30

providers need that amount

1:12:32

of information every day to milk

1:12:34

that information, to make

1:12:37

money, to sustain additional growth and

1:12:39

new data centers. It's like, you

1:12:42

know, a circle that goes around and

1:12:44

keeps growing. It's like, and it's

1:12:47

feeding itself. Without us

1:12:49

sharing more information, more photos daily, you

1:12:51

know, they will never survive. The

1:12:54

cost of maintaining their infrastructure will

1:12:57

crush them. And

1:12:59

so this long story to say

1:13:01

that there isn't why we created holepunch and

1:13:03

kit is to showcase

1:13:05

that, you know, this interview could have

1:13:07

been done through kit, completely beard to

1:13:09

beard without any central server. If

1:13:11

holepunch, the company that is building kit, would

1:13:14

die tomorrow, kit will keep working. Imagine

1:13:17

if you are in a country, right? So

1:13:20

kit is serverless, allows you to

1:13:22

do a chat. We have, we created

1:13:24

a Bitcoin chat on kit with,

1:13:27

there are 1.5 thousand people today. There

1:13:29

is no central server. They are talking real

1:13:31

time. They're sharing files. You

1:13:34

can share any size of

1:13:36

file you want because there is no limit.

1:13:38

It's just your bandwidth. You're interacting with each

1:13:40

other. You don't need, if

1:13:42

you share a file on Zoom,

1:13:44

you are limited to one other megabyte because

1:13:46

otherwise if everyone was sharing two big files,

1:13:49

then you would crash, you

1:13:52

know, Telegram or Zoom or whatever, right? But

1:13:55

when peer to peer, you don't have those limitations.

1:13:57

So what we wanted to create with kit on

1:13:59

holepunch, is we wanted to

1:14:01

prove to the world and to all the

1:14:03

developers that you can build peer-to-peer applications with

1:14:06

a great user experience without having

1:14:08

to have sent out servers. And

1:14:11

so, KitLao is being used by

1:14:13

many people, have much

1:14:15

better video quality, much better

1:14:17

audio quality just because if

1:14:20

you are talking to a person with living in

1:14:22

the same area, it

1:14:24

feels like you have that person in front of you.

1:14:27

Because the latency is the

1:14:29

smallest it can ever be because the

1:14:31

messages, the data packets will find

1:14:33

the shortest path between you and

1:14:35

that person. And that between,

1:14:38

of course, you, the two devices. It's

1:14:41

not magic. The internet was built in

1:14:43

that way, but there was never

1:14:45

an incentive by anyone to

1:14:48

build it, to use it in that way because

1:14:51

the companies that had

1:14:54

the most money had the incentive to

1:14:56

make it centralized. And all

1:14:58

the rest, all the nerds that say like

1:15:00

us, they never had the

1:15:02

money to build highly

1:15:06

successful applications to

1:15:08

prove that internet can

1:15:10

become server-free in a

1:15:13

way for, or at least most of it can

1:15:15

become server-free. And so,

1:15:17

that, long story short, Tether

1:15:19

found recently, as we chatted a

1:15:21

little bit about it, with

1:15:25

an important profitability.

1:15:28

So since we are, as I said,

1:15:30

we are in for the technology, we

1:15:32

wanted to create the

1:15:34

freest, free as in freedom, chat

1:15:38

solution that is

1:15:40

unstoppable, that is private and doesn't

1:15:43

need any central server. Super

1:15:46

cool. So I get

1:15:48

keyed is kind of like basically

1:15:51

decentralized peer-to-peer, alternative to like sort

1:15:53

of WhatsApp and Telegram. Do

1:15:56

you see the same protocol?

1:16:00

Also as a foundation for

1:16:03

very different types of applications?

1:16:07

Sure. On the same protocols you

1:16:09

can build peer-to-peer Uber.

1:16:12

One of the scariest things is, imagine that

1:16:15

this baby comes where you

1:16:18

monitor your children. They are

1:16:21

all the data passed through a central server that

1:16:23

you don't own. How scary is that?

1:16:26

You should be able to connect directly from

1:16:28

your phone to your home device without needing

1:16:31

crazy configurations. Oh,

1:16:34

that is possible through

1:16:36

holepunch. You

1:16:38

can build peer-to-peer mapping.

1:16:42

Whatever you think is built today, you can

1:16:45

build a holepunch. When

1:16:47

I talk to other developers, I

1:16:50

tell them I shook his holepunch.

1:16:53

I shook his, I can build interaction

1:16:57

between two devices without any middle

1:16:59

server. One device provides

1:17:01

some services to another device without

1:17:04

both phones, maybe, without

1:17:06

any central server. Even

1:17:09

developers are mind-blown. That

1:17:12

tells a lot about our education. As

1:17:14

developers, when we go to the university,

1:17:17

we are told that

1:17:19

the client-server model is the

1:17:21

only model existing. You always need

1:17:24

a server for many clients. That's

1:17:26

not true. We wanted

1:17:28

to prove it and we are going to invest

1:17:30

a lot of money to make sure that education

1:17:32

will pick up this new partner because we want

1:17:34

a horde of developers

1:17:37

to build peer-to-peer applications

1:17:39

that has decentralized applications

1:17:41

to reduce the dominance

1:17:44

of the military big

1:17:46

tech companies. I

1:17:49

hear all of that and I'm a big

1:17:51

fan. I will definitely check out holepunch. One

1:17:55

thing that always got me

1:17:57

about these peer-to-peer platforms like

1:18:00

Bittorrent was that there was

1:18:02

an incentive layer missing. So

1:18:05

kind of people basically was

1:18:07

very sensitive to kind of people who would

1:18:10

exploit it by kind of downloading

1:18:12

and not seeding. So in

1:18:15

principle kind of like on any crypto

1:18:17

infrastructure you can kind of build in

1:18:21

an incentive layer. Is that happening with

1:18:23

HoldPunch as well? No,

1:18:25

not really. So I mean, HoldPunch

1:18:27

is based protocols. It doesn't have a token.

1:18:29

We'll never have a token because I believe

1:18:32

that is a simple protocol that can, well,

1:18:34

it's quite complex, but it's a protocol that

1:18:36

can be taken by everyone and you can

1:18:38

build whatever you want on it. Now,

1:18:41

I agree that there were with

1:18:44

Bittorrent, most of the people wanted

1:18:46

only to download but never give up their upload

1:18:49

bandwidth. That's completely

1:18:51

true. But let's think about the other

1:18:53

side, right? If you are building like

1:18:55

a peer-to-peer chat and

1:18:57

you want to use a chat, your incentive is

1:18:59

to talk to people. So

1:19:02

there is no actual cost for you.

1:19:04

It's not that you are running a

1:19:06

node that allows people to run their

1:19:08

traffic through you. So

1:19:12

the difference between HoldPunch

1:19:14

and a blockchain is that in a blockchain you have

1:19:17

a global share state. So if you run a node,

1:19:19

you will see all the traffic, you will see the

1:19:21

mempool data and so on. With

1:19:23

the HoldPunch, you only choose to

1:19:25

be able to talk and to connect to

1:19:27

the people you want to because you are

1:19:29

creating a network with them. So

1:19:32

HoldPunch is not a huge big network, but

1:19:34

it's a protocol to create your own small

1:19:37

network with all of the people you want to talk to because

1:19:39

you are interacting with them without any

1:19:42

central server. That's the only

1:19:44

way to make something scalable. Blockchains need

1:19:46

to use a global share state because you

1:19:48

need to make sure that no one is

1:19:50

double spending. But

1:19:52

you don't need that. If

1:19:56

you are creating a communication system, you

1:19:58

don't want to receive messages from

1:20:00

someone from the other side of the world to

1:20:02

route them somewhere else, right? So it doesn't matter.

1:20:05

You shouldn't be a starter yourself. There

1:20:07

is now this other program

1:20:09

project called Noster. They

1:20:13

use relays, centralized relays. I always

1:20:15

make the comparison between Noster and

1:20:17

Tit. As in, with

1:20:20

Noster you have relays, so you need to

1:20:22

provide incentives to people to run relays, otherwise

1:20:24

nothing will work. With holepunch

1:20:27

you don't need relays, so hence you

1:20:29

have a simpler, more simplified network. And

1:20:32

you don't have to have incentives, because the incentive of

1:20:34

people is already talking to each other. For

1:20:36

example, now you are in... Imagine

1:20:38

that you were in UAE, in

1:20:41

Dubai. You cannot use Telegram or

1:20:43

WhatsApp to call your friends and

1:20:45

family, because, you know,

1:20:47

the Telegram and WhatsApp are blocked. You can

1:20:49

use Kit to talk to anyone, and from

1:20:52

there and to there. Because

1:20:54

the beauty of

1:20:56

Kit is, because it's peer-to-peer, in

1:20:59

order to block Kit you will need to block

1:21:01

the entire Internet. Because

1:21:03

it's not predictable where you connect to. Super

1:21:07

interesting. So is the underlying technologies

1:21:09

in some sort of whisper

1:21:11

network, or how does it work? It's

1:21:14

used the same concept of BitTorrent. In order

1:21:16

to find each other, we use

1:21:18

the same concept of BitTorrent called a DHD,

1:21:20

the distributed hash table. So

1:21:22

that the

1:21:25

distributed hash table is in

1:21:27

BitTorrent, around 10 million nodes in

1:21:30

size at the peak. So you have

1:21:32

all these 10 million nodes that are acting as... Not

1:21:36

connection points, but are temporary

1:21:39

databases. It's

1:21:42

basically, they are key value stores that are

1:21:44

storing part of the global index in

1:21:46

a really resilient way. So

1:21:49

if you shut

1:21:51

down a part of the network, even 1

1:21:53

million nodes, the network will keep surviving and

1:21:55

will adapt to the new set of nodes.

1:21:57

So it's really the most sophisticated.

1:22:00

created, resilient,

1:22:03

distributed database ever created, and we

1:22:05

used it in order to allow

1:22:07

people to find each other through

1:22:10

the four key. Because if

1:22:12

we added a centralized index to key, that

1:22:14

would be centralized. The

1:22:17

point of centralization is that you cannot

1:22:19

just add just a little bit of

1:22:22

centralization. Right? So either

1:22:24

you are centralized or not. There

1:22:26

is no in between, because the moment you are a bit

1:22:28

centralized, then you are centralized. I'm

1:22:31

curious, so one of the things, one of

1:22:33

the projects I've been pretty deeply involved in

1:22:35

for quite a while is a thing called

1:22:37

Urbit, which sort

1:22:39

of envisions basically different

1:22:42

architecture for computing and an alternative

1:22:45

to the internet where basically all

1:22:47

of the computers work in a

1:22:50

kind of P2P array where applications

1:22:52

run locally. Have

1:22:54

you looked at Urbit? Do you have any thoughts on the

1:22:56

sort of trade-offs between

1:22:59

the whole punch approach versus

1:23:01

that? So

1:23:05

I really like Urbit. I

1:23:07

mean, I played with it I think

1:23:09

in 2017, 2018. I

1:23:13

don't remember when, but I think it's a

1:23:15

really not old project in a

1:23:17

sense of old, but it is around since a long

1:23:19

time. And the

1:23:22

concept is great. I think,

1:23:25

and the way it is designed also, I

1:23:27

really love it. I

1:23:29

think Whole Punch is creating protocols for the

1:23:31

real world, right? Doesn't

1:23:34

try to change the world completely starting

1:23:36

from scratch. We already

1:23:38

have internet. We have

1:23:41

immediate needs that are creating a

1:23:44

layer on top of internet that is secure, that

1:23:47

gives sovereignty to people. And

1:23:49

we are doing it in the most simple way, in

1:23:52

the most modular way, because with Urbit you had, with

1:23:54

kind of the way I felt

1:23:57

it is a monolithic approach. You

1:23:59

have to take it. all and everything should run on

1:24:01

orbit. Otherwise it would, you know, what

1:24:04

wouldn't work. But with wholepunch is

1:24:07

we have more than 100 GitHub

1:24:10

projects and libraries. You can

1:24:12

take pieces and

1:24:14

use it in craft it for an

1:24:17

integrating in your real application, your existing

1:24:19

application already. Right. So we

1:24:22

are taking a different approach in order

1:24:24

to change the internet. That is giving

1:24:26

you the tools to do small steps

1:24:28

toward the centralization rather than forcing everyone

1:24:30

in a new, completely different

1:24:33

ecosystem. Yeah,

1:24:35

no, absolutely. I absolutely, I think

1:24:37

that's a fair description in

1:24:40

terms of the differences. Do

1:24:43

you see Keats becoming

1:24:45

some kind of business down the

1:24:47

line or is it just a

1:24:49

sort of public service that you guys want to

1:24:51

develop? I

1:24:53

think it will remain

1:24:57

public service. I mean, the

1:24:59

point of Keats is not making money.

1:25:01

You know, there is this, sorry, making

1:25:03

many references to memes, but well,

1:25:06

you know, the, from Batman Joker is

1:25:08

not that he said, it's

1:25:10

not about the

1:25:12

money is about sending message and

1:25:16

for us, Keats is that, right? So we make

1:25:18

good money with Tether. We, it's not

1:25:20

that we have to make money every single time

1:25:23

we do something. Sometimes it's

1:25:25

important to send a message. And

1:25:27

there is a pun intended because Keats sends messages. And

1:25:31

I think this is a fantastic place to end

1:25:33

because it is, it is such a

1:25:35

like, you know, message

1:25:38

to go out on. I

1:25:40

will definitely explore Keats and Whole Punch more.

1:25:43

I think the entire interview was

1:25:45

super fascinating, but somehow the whole punch part

1:25:48

got me most as kind of like a

1:25:50

peer to peer and decentralization maxi. If

1:25:53

people kind of want to learn more about Whole

1:25:55

Punch or indeed Binfina XO

1:25:57

Tether, where can they go to? Where can

1:25:59

we send them? So,

1:26:03

on X there is, until we replace

1:26:06

X with something built on whole punch,

1:26:08

so just a disclaimer.

1:26:11

So, you could go on at

1:26:13

Tether underscore TO or at

1:26:16

Beatfinex or at allpunch

1:26:19

underscore TO or at kit

1:26:21

underscore IO or at

1:26:23

Paolo, Arduino, so name is your name

1:26:26

to find all my memes and something

1:26:28

about everything that I do. Cool.

1:26:31

Thanks so much for coming on Paolo.

1:26:33

Really enjoyed the conversation. Thank

1:26:36

you guys. It was

1:26:38

super fun. Thank you for

1:26:40

joining us on this week's episode. We release

1:26:42

new episodes every week. You can

1:26:44

find and subscribe to the show on

1:26:46

iTunes, Spotify, YouTube, SoundCloud, or wherever you

1:26:48

listen to podcasts. And if you

1:26:50

have a Google Home or Alexa device, you can tell it

1:26:52

to listen to the latest episode of the Epicenter podcast. Go

1:26:55

to epicenter.tv slash subscribe for a full list of

1:26:58

places where you can watch and listen. And

1:27:00

while you're there, be sure to sign up for the newsletter

1:27:02

so you get new episodes in your inbox as they're released.

1:27:05

If you want to interact with us, the guests, or

1:27:08

other podcast listeners, you can follow us on Twitter. And

1:27:10

please leave us a review on iTunes. It helps people

1:27:12

find the show and we're always happy to read it. So,

1:27:15

thanks so much and we look forward to being back next

1:27:17

week.