Martin El-Khouri on building an economy for machines

0:00We are going to see that huge transition from software builders to hardware simply because software is solved, right? There's not a lot that you can do. It's it's it's a it's really difficult to build a mode solely around software. Hardware is different. I'm Ashan Addison from the Crypto Coin Show and today on blockchain interviews we have Martin El-Khouri, chief of business

0:27development at Peaq, a layer 1 blockchain built for the machine economy. Decentralized physical infrastructure networks, also known as DPIN, letting real real world machines like robots, electric vehicles, and sensors get onchain identities, get paid, raise financing, and much more. Peaq has had one of the busiest stretches yet. A new chain launch on

0:51Salona and a omni chain strategy, a new tokconomics model, real industrial robot partnerships, regulatory in Dubai, all within the last month. There's so much to talk about. Martin, welcome to the show and thanks for taking the time.

Thanks for having me, Ashton. It's a pleasure.

Yeah, it's my pleasure. I've been covering peaq for over 5 years and

1:14starting in 2021. It was so early and we're still so early in the machine economy, but now it's really starting to kick up compared to then. And especially in the last month, a lot of peaq developments happening, expanding as more robots and machines come online and get connected into AI and the blockchain. I would love to start out with your take in one breath. What is

1:39the bet that Peaq is making with machines like robots and electric vehicles?

God, it's going to be very difficult to be brief on this one, right? Listen. So, when when when Peaq started out, the underlying assumption has always been that there is an economic force that is by orders of magnitude larger than humans. And that is obviously

2:05Physical bodies that are enabled by by artificial intelligence by agents that do work autonomously that are able to do multiple tasks at the same time and that are basically creating and recycling value 24/7 and that is the machine economy. The machine economy is a new species of physical hardware. It can be sensor networks. It can be

2:30different types of devices. It can be drones, cars, airplanes, humanoid robots, service robots, collaborative robots, whatever whatever it is that is actually participating in the economy will do so more and more autonomously. And that value creation that we will witness and we're seeing like the first innings of that already happening and unfolding in front of our eyes that

2:55will increase the global economic output by again by orders that we can't even imagine. And so in order to enable that economic layer, some fundamental pieces have always been missing. And this is where we started out. In the very first cases that we did when we started we were working very closely with some of the leading enterprises in Europe.

3:16still obviously have great relationships, ongoing active partnerships and the work that we did when we started like 9 years ago basically informed everything that we're doing today and kind of catered to how Peaq looks and evolved into this omni chain coordination layer for robots and machines to basically do business and we don't want to

3:40own the payment rails for these machines. There is stakeholders that do that very very well. We basically consider ourselves as again the enabler right onchain identities omni chain composable wallets access control for robots. So basically POS which is the stack right that is kind of a consequence of the last 9 years of work is really yeah kind of the

4:06one-stop shop that enables robots and machines to then not only walk around and navigate and do all the stuff that other people work on and focus on but to actually be able to do business, make money, sell their services, consume services and most importantly be investable financial participants. Definitely. And I'd love to dive into

4:28that part about the onchain identity and the wallet for we're already seeing this with AI agents and soon to be humanoid robots. I've seen the alphas for them. I'm ready for them to clean my house. But how does that work with

how does that work with the them needing an identity? Why do they need it? And the wallet?

Yeah, very simple, right? So if you

4:54think about how our civilization has developed, there were some things that nobody really talks about that were paramount to us being able to do what we do, right? So I always like to use simply the simple example of you buying a house, right? If you want to buy any sort of real estate, obviously you can't do that without having a p like an identity, right? You

5:16need your passport or your identity card in order to verify and authenticate that you are who you claim to be. You obviously need a bank account to be able to pay the person that is or the broker that is selling the house, right? You need a credit score, some some sort of a creditworthiness so that when you borrow money or when you when you lend money from the from the bank

5:40that the bank is able to prove that you are or to verify that you will be able to repay that loan over a predefined period of time with a predefined set of interests. And there needs but they need to be we need to have options that are basically triggered similar to how you have that with a bank where you have securities you give something as a collateral.

6:03At the end of the day all of these things are super boring and not not exciting. But this is what enabled at a very fundamental level what enabled globalization to happen and what basically enabled let's say the civilization that we live in to actually happen and emerge. Crossber flows of capital at the end of the day always require identity checks. They

6:26always require financial institutions that facilitate these payments. They always require a certain degree of accountability. And that basic stack that we have as humans just does just does not exist for machines. And that's and that's interesting because in stark contracts contrast to AI agents with physical hardware you have an actual physical value that is always

6:52there as an underlying collateral or could be used as an under underlying collateral. And so basically when we started working on the identity for us really the case was was not so much about the financialization side of things like the biggest eye opener was when we when we did a big case or big partnership actually back in the days with the Volkswagen group and Audi

7:14there the identity needed was was needed in order to harmonize two different systems right within the group. So an Audi car being able to charge and pay a Porsche charging station for the charging session. So it was a very pract practical need that we discovered and where we said okay if you think about how this works with current enterprises then technically the way

7:37machines authenticate themselves are basically like passports that are only let's say accepted within one country so completely useless right and so you need an open system and you need an open standard that you can plug into and this is basically what the peaq ID is and everything else is then tied to that ID so what that enables is similar to how we have certain

7:57standards for payments as humans. How we basically have our identity cards. Different countries accept different passports or don't accept different passports. But we have some sort of a harmonized system and that is what we build around the peaq ID for these machines. And the big advantage and the importance of that is that this vision of the machine economy where you

8:18have hundreds of billions trillions of machines and sensors and devices communicating with one another, verifying each other, transacting with one another. I mean the way the world looks right now that's comp like completely impossible simply because the systems are not compatible and that is what we're solving and that is why I always like to say in contrast to

8:40many other verticals or niches within within our industry within web 3 you do not need to architect a product market fit for blockchain and machines. It is a very natural thing because there is no no real alternative for machines to basically have an open system. And then the second thing the way machines are built these days specifically robots is one huge

9:06advantage is that a lot of these robots always also are operatable in an open source system. So for robotics it's it's it's the robotics operating system ROSS. And that basically already gives quite a big pool of machines that automatically are basically composable. So you do not even need to get into the proprietary software stack of these

9:28of these machines. And so everything that we build, everything that we determined is kind of derived from what is needed for humans to do business and translated to machines, physical machines, robots, devices. And yeah, that kind of helped us to then initially bootstrap peaq as an as a layer 1 with a clear focus to you know onboard all of these sensors and devices. But around

9:52this time a bit earlier like a year one and a half years ago we already started kind of thinking about how we could be more efficient, how we could leverage the core value proposition of what we do more strongly. And there then one kind of coralization was and I had this example with like a passport that's only accepted in one country. If you basically exclude all other

10:16chains and all other ecosystems, it's a it's not smart because it kind of hampers your adoption, right? And b it's not necessary. It kind of undercuts the entire entire value proposition of having that. And so we said look let's let's let's basically make that stack entirely omni chain and focus it on POS. The chain itself has a huge value proposition. It is the neutral

10:40ground where where the ids live that never changes. But if a machine wants to do business on Solana tap into the liquidity on Solana or other chains it should be able to do that because at the end of the day for us our first class citizens are robots and machines.

Definitely and I agree with that. I think the industry has come to a

11:02conclusion over the last 2 years that we needed to aggregate liquidity as much as possible and bring harmony between different blockchains and communication. So the POS launched on Salana not not too long ago and 23 other chains I believe. I'm curious if there's value in all 23 chains if that's extra work for the team to integrate,

11:27but can you talk about Salana primarily as probably the number one go-to place for machines? I know it's fast and cheap. But what does this enable for people and machines?

Yeah. I mean to tap into your first question right like enabling any EVM compatible chain is pretty pretty straightforward right of course it's it's some work that needs to be done but

11:50it's pretty straightforward obviously with Solana it was a bit of a bigger effort but the team has been super supportive throughout the process and yeah it works anywhere so what is the rational that we take right so when we launch the layer 1 obviously layer 1 blockchain is the technology it's one thing right like for us always kind of the key differentiator was the

12:16Proprietary stack that is purpose-built for these machines and then basically enables all of these machines to also be atomically composable basically that's that's one part you know but at the end of the day you need access to capital at scale at a very large scale you need distribution even for robots and machines you need distribution If you

12:40want to raise funds for a machine, you need distribution. If you want a machine to be able to monetize its services, it needs distribution. And even those that build the machines, and we will probably speak a bit about what we call the machine money markets and our visions for that, right? Bootstrapping that is incredibly hard. And when we're speaking specifically

13:01about Solana, Solana to me is by far the best when it comes to providing the system for capital formation. Providing that access to capital in a way that is that caters to the needs of different profiles. Solana has a sensational distribution and so and then plus also something that is very much worth considering Solana already has

13:29quite a strong number of quite impressive and strong deepens that are already building within the ecosystem right so it's a very natural fit we were very fortunate to also through mutual partners that we that we work with in different part of the world kind of have have a kind of direct direct a direct channel and a direct communication. And when we really started speaking, it just

13:53felt like a very obvious match. Because the chain itself provides everything that is needed in order to let such a machine economy happen, which which obviously always should be a prerequisite for any sort of partnership to happen. But then most importantly, there was just a genuine interest in what we do. And so that's why we basically not only expanded the token to

14:17Solana but also launched Picos on Solana and have the first machines activated on Solana. We will build build the machine money markets also on Solana. There will be there will be very very close alignment and yeah that's that's really the rational behind it.

Well congrats on the expansion. I'm looking forward to seeing how it plays out. And with the growth of machines

14:44onto the network, you have a public dashboard that shows the machines activated, the economic value. Right now, it's showing over 2.5 million machines, which sounds like a great number, but I'm sure there's probably billions or more machines in the world. Can you talk about the progress so far? What does that number mean to you? Well, so

15:13with this new approach, right, and it's it's very important to not only speak about the machines in isolation, right? I think it's very important to understand how the new economics work, right? So in the new economic in the old economic model, it was like let's say the traditional layer one economic model, but what we what I just described with building the coordination layer,

15:33building the trust system basically for robots and machines to do business. What that means is that we are introducing an entirely new token utility that to me is the perfect to the perfect fit for what we do and is something that also is very replicable for any other builders within within the hardware space really not not only robots and machines and that is the

15:54token bonding right so basically what is happening is that every machine that is activated and joins the network regardless on which network is bonding peaq there are different tiers these tiers is are denominated in different prices in stable coins, right? and then oracle stamps. What that translates to in terms of peaq tokens that these amounts

16:19of these peaq tokens are then bonded to the individual machine for one year and the machine start to build what we call economic weight within the system right and then if the depending on the tier different services of POS can be accessed or are included so they can always be of course upgraded and then what happens is that a year after these

16:43machines are starting the subscription of course they will decide right or the operators will decide do we want to continue do we want to not continue if they continue then obviously they consume more peaq bond more peaq builds stronger economic weight if they don't then half of this token bond is burnt and the other half returns to the

17:06treasury right and so that mechanism and you will you will see how that unfolds also also next year specifically Typically when we're rolling out also the entire trust validation system with all the different trust validators at scale that economic weight is obviously important because it's it's reputation at stake to a certain degree for that individual

17:26machine. So to go back to your question why does that machine count matter? I think it matters for multiple reasons. A it's kind of an indicator for adoption. To be intellectually honest also we communicated that very transparently. Many of these machines are legacy machines right machines that we already had within the ecosystem where the projects were were keen to

17:51also participate in the new economic system in order to be able to leverage the value proposition that Vos now provides. But we are onboarding new machines every single day. And what that really means is that from a token perspective, obviously it's it's it's translating the revenues that Picos generates directly to the token, which which is fantastic and

18:16very important, but more importantly, it gives us an idea on the actual product market fit for Picos. And here I mean we I have to say I mean I was pretty surprised specifically with this dujan partnership that we have entered where we are by it's structured across four phases. We're basically now in phase three. I personally expected that this will take

18:39much longer. But the demand from some of the leading the leading players in the context of machine financing is breathtaking right and obviously having these all of these machines being onboarded all of these machines being financable and most importantly as you will see very soon the implications of these machine identities of these new economic

19:04participants and the opportunities that they will get through the machine money markets is what we hope to be kind of the birth birth pangs of a birthbangs of a new economy, right? And so really for us it's the number one metric that matters. The number one metric that we look at every morning when we open our computer every night when we go to sleep is that machine

19:26count. Of course, it will not always, you know, go straight up because

there is big BD efforts still that are that are needed in order to

to onboard more and more machines. But the goal is and I mean it's a very ambitious goal but the goal is to be the de facto standard the de facto financial standard for the machine economy and that is what we what

19:52we're building it for. And so look like there's robots we have robots on the system we have deepen devices we have cars we have all different sorts of things and more and more of these things also coming in the next couple of weeks which will be also quite exciting I think. But there's also there's also sensors there's smaller devices right so

I think millions is a small number

20:16for what yeah for what we should be able to achieve right so we want to within the next years obviously the goal is to get into the billions

yeah and that's really exciting and I would love to dive into that dujon partnership about industrial robots I believe in a Second, I'm curious on these different types of machines. You know, obviously there's everything from

20:44from sensors to cars and eventually humanoid robots and all the other

computers and devices in your house.

What's the what's the lowhanging fruit? And you know, I know there's going to be a lot of beady efforts and major partnerships like this Djon partnership, but how do you get to 50 million or 100 million and the next big milestone? Yeah, I mean right now the BD efforts

21:09are not so much focused around like ramping up the number. They are very much focused around enabling the machine financing side, right? And so obviously if you want to deploy a thousand sensors, you could probably do that, right? Like it's not not super expensive, right? But where we see big product market for this collaborative robotics, which is not

21:31something that we were expecting expecting, right? So to give you and the listeners a bit of context. So automation has been along around for a very very long time, right? KUKA, AB, like all of these big robotics companies, they didn't just like emerge within the last in the last couple of years, right? They've been around for decades. What we need to understand

21:56is that usually if you've ever worked in a I did actually a student job in Breman in Germany for Mercedes where I was like driving the forklift around and so I was within within the factory during the summer break and there I saw that there is very clear areas that you cannot enter and that is where like the actual assembly is happening with all of

22:18these robots like no one can interfere with that. Why? because these types of machines, they don't have a stop function. Like if you're in the way, you're in the way and you're basically dead, right? So the first iteration and that came up just a couple of years ago actually. It's not it's not a super old development like five six years ago. It really started with these

22:38collaborative robots which basically are built so that humans and robots can work within the same space, right? Technically these robots are super performant you know. The reality though is that adoption has been hampered a bit for two reasons. One they cannot work at the maximum performance for the very simple reason that you need

23:04to be very careful with the security constraints and everything that is that is happening around that. But then secondly, the financing, right? Because even though they are much cheaper than than let's say the large scale industry automation offerings, they are still relatively expensive, right? And you're not basically just buying like one one robo arm and deploying it when you

23:26actually automate your production as an you will probably have to spend somewhere between 500,000 and 3 4 5 million in order to get to that point. And the big issue for all of these machines, that's also true for service robots, that's true for drones, that's true for all of these things where we know it's going to happen, we just don't know when, is that in stark contrast to the

23:50initial example that I gave for gave for us humans, real estate, etc. You have no idea about the residual value, right? There's literally no understanding. And that is also the case within within the collaborative robotics. So the reason why we see a big product market fit already within that specific field is because we basically can get

24:14anonymized telemetry and operational data for these machines. We can get them at scale. We can get them in a way that is secure and protects the IP of those that actually deploy the machines, right? So there's not a lot of risk and we are able to basically finance these machines against themselves right and that is a very very big value proposition that it exists right

24:42don't want to say that asset back financing for machines and hardware doesn't exist but it's a huge mess right it's very traditional very slow very cost inefficient and that is really where poss is a huge door opener right And the more data a manufacturer provide pro provides the better terms we can offer on the financing. Plus of course with the other modules

25:05that POS has has is obviously the machine can monetize some of its resources while it's laying idle. While that's not the core focus right now, it's still one of the well one of the value propositions. And so that is that is a vertical that we are really focusing on because it's machines that are used already. It's not a humanoid, you know, that basically God knows when

25:28when they will be able to stand right next to you in your kitchen and kill you with a knife.

It will happen. Hopefully, they won't kill you, but it will happen.

but the actual business opportunity in collaborative robotics, in service robotics, which is another se sector that we're focusing on, and also in transportation, logistics

25:48and mobility in general. There we see a big product market fit and there will be more updates within the next couple of weeks where you see how that also translates into into commercial partnerships and opportunities.

Very interesting. Looking looking forward to those. You know, I did see a video of this humanoid robot. I think it was in like a nail salon

26:12

working and then somebody pushed it and all of a sudden the Bruce Lee programming kicked in and it tried to kick this guy. You know and beat him up. So maybe that's a security issue that would that be something a security issue that machines on Peaq OS, you know, Peaq might have to deal with or maybe more relatably messing with the wallets or the identity and trying to

26:39fake the payment or the identity and things like that.

Yes. And that is exactly why Pico exists, right? So basically whenever something like that happen happens the machine will be slashed right and it will be basically it will for instance let's not take a kickboxing humanoid as an example but another example that I always always like to

27:00give right I don't know if you guys heard of this drone incident that happened a couple of weeks back in Germany and at the airport in light it's actually a perfect example so there was this drone nobody really know knows where it cames from there. It came from there is some allegations but I don't want to go go there. And it was equipped with explosives landed right

27:25next to some sort of a I think it transport jet from the that was supposed to fly to Ukraine or whatever. It didn't explode, right? But for these long range travels, assuming it was traveling for quite a while, right? Like drones need to charge. So,

there is obviously providers that build these charging stations. We also working with some of these and obviously

27:52infrastructure needs to be open or as open as possible, right? Because if you if you're Uber and you want to offer drone delivery, you are happy to pay for the drone, but you are not happy to equip the entire country with a drone charging infrastructure that everybody else is also benefiting from, right?

So there's that's where the financing and POS comes in on the one hand, but

28:11then most importantly and that is where this example comes in. You need to be 100% sure that the drone that is using that infrastructure which is critical infrastructure is allowed to actually use it. So it needs to authenticate itself, identify itself with the respective charging station. It needs to be allowed to charge on that charging station. And that is why specifically

28:36for these higher risk topics, right? What is very important is that you have hardware at a station and verification. And that is why we I think last week we introduced Pico verify which is basically chip level signature, right? So basically the machine signs on the secure element within the machine. So authenticates itself and verifies that it's that it's real.

29:04it's basically as close as you can get to 100% security. I mean obviously some if someone would would basically steal the drone, right? And manipulate it in some shape or form or put that secure element somewhere else then it can still be fake. But that is why we have like several verification layers, right? And so if a drone is basically leaving operations for

29:26whatever reason, then obviously the operator will report it and then it's the different like the oper operator level verification and the POS verify function basically aggregate all of these data and make that visible. But the most important part is that if the machine is not actually entitled to do something it should not be able to do that. And that is exactly why that POS

29:47stack and the identities of the individual machines, the machines having the wallets because that is also true for payments, right? Like a machine needs to be able to pay decentralized infrastructure autonomously. That cannot yeah that cannot happen when you when you basically don't have these security layers. But we can obviously not prevent the machine from going rogue,

30:09right? Like that is that is that is that's the job of other people. That's on the intelligence layer, right? And we're not touching the intelligence layer.

Yeah. Yeah. You mentioned at the beginning building the foundations, the identity, the wallet, the, you know, reference as, you know, the boring but necessary stuff. What are you most excited about

30:31In things that might get people really interested in the convergence between robotics and blockchain over the next 12 months that Peaq is getting involved in?

Yeah. Yeah, like one of our core focus topics is what we always talk about or tease, you know, on our personal X channel, sometimes even through through the main account is

30:55what we call machine money markets, right? And so we've we've had the first iterations of that. Obviously initial machine offerings that we do with partners like Dulment, with Coin List, with others is one example, right? you being able to basically finance a fleet of machines directly through through onchain capital. We already like teased our collaboration

31:22with machinef fun so people know that this is coming which is basically the world's first hardware builder launchpad. It's it's basically fair launch launchpad. Any project can can launch there. We can't really prevent it. So you might see memes, you might see some some random stuff. But the reason why we built it is because we attached the pico sag to

31:45the launchpad. And we also attach Kodak flow simarina who we also partnering with in that contact context so that builders can basically upload their simulations when they have a prototype they can connect the prototype to POS. They can prove that they have that machine. they can showcase that there is actual development going on and basically our vision for that is

32:07that the these token launches like fair launches, pump fun launches, ston launches, right? Yes, a lot of it is memes, but some projects actually manage to get from nothing to first revenues and first customers and codec who we're partnering with in that context is one of these examples, right? So we're basically trying to replicate that and in an ideal world we'll be able to

32:31yeah see the first unitary funded out of out of machine fun or a similar project like unitry in a couple of years you never know from nothing to a billion dollar company and it's not it's not impossible. It's it's unlikely but it's not impossible because at the end of the day we are going to see that huge transition from software builders to

32:52hardware simply because software is solved right there's not a lot that you can do it's it's it's a it's really difficult to build a mode solely around software hardware is different right so there is a lot of applications it's hyper local which is also something that people need to understand right like you have if you have a fleet of cleaning robots in Cincinnati then this fleet of

33:14cleaning robots will not be able to drive all the way to Hong Kong. Right? So, you'll have similar types of hardware businesses emerging all across the world, profitable companies, people that basically just start out and that's something that we're very excited about. And then we have two more. So, we always internally that's alpha actually. We're always speaking about we're

33:34always speaking about four rails. The IMOS are one machine. fund is one and then two other rails which are large scale financing products, industrial financing products where I cannot really speak a lot about but what I can say is in the context of Dujan that is already tested being tested as we speak actually. So there will be the first commercial partnership hopefully

33:57hopefully soon. Let's see. And then yield products. Yield products that basically allow new ways of hardware builders, deepen builders to access capital without let's say the traditional constraints that you as a builder have but also that you as an investor have. That's also being piloted at the moment. So there will be these these four different types of machine

34:20money markets and we are planning to roll all of them out within the next quarter hopefully. You never know at scale probably it's going to take a bit longer but you will learn about all of them. So I'm really excited for this convergence.

Yeah. No sounds really exciting and what I what I've taken away from that as you said is you know software has been

34:44solved. The smart people are now working on hardware. There's there's so much to be done with regards to robotics tying in the software, but we need that hardware if we're going to move the world forward.

Absolutely. Yeah, I agree.

Incredible, Martin. Well, thank you for the time. What's the best way to follow along with these updates and these

35:04different rails that you've mentioned here at the end?

Yeah, sure. I mean, obviously our X account is always a great great source of updates. There are different telegram groups depending on where where you guys are at community groups but I would always recommend follow us on X follow also the team members right the founders for

35:26instance myself we are very active we always also like to speak about give some more background information about about things that are going on so that's one thing and obviously feel free to reach out if there is anything most importantly if you're building in hardware if you have an idea it doesn't need to be something that you like believe it will change the world

35:45overnight. It can be just an idea like a small little robot that you want to build, reach out. We want to we want to see where we can help. Very exciting. I will leave some show notes with the links to the platform and the socials as well. There's so much to continue building and I would love to follow up again in the near future because I know there's going to

36:07be major announcements from Peaq. Thank you so much for the time, Martin. Thank you for having us, Ashton. Big pleasure.