Inside the AI Revolution: AGI, Decentralized Intelligence & the Machine Economy

Live from Proof of Talk Paris 2025, The Edge of Show dives into the real-world rise of Decentralized AI Agents—the shift from chat to action. Host Josh Kriger sits down with Humayun Sheikh (CEO, Fetch.ai; Chair, Artificial Superintelligence Alliance) and Janet Adams (Co-founder, Artificial Superintelligence Alliance; SingularityNet), then explores Deepin (decentralized physical infrastructure networks) with Leonard Dorlöchter and Till Wendler of Peak. From the launch of ASI One—an agentic LLM that can discover, talk to, and coordinate with other agents—to SingularityNet’s decentralized AGI ambitions and Peak’s blueprint for a machine economy where EV chargers, robots, and city sensors transact autonomously, this episode makes the stakes clear: in a world moving “AI-first,” Decentralized AI Agents become teammates, marketplaces, and infrastructure all at once. You’ll hear why agentic systems could erode aggregation monopolies, how SMEs can adopt AI in “two clicks,” why decentralized stacks may outpace centralized models, and where Deepin meets smart-city deployment, machine-RWA, and Decentralized AI Agents coordinating real economic work. If you’re building at the intersection of AI, crypto, NFTs/Ordinals, and DeFi, this conversation is a playbook for what’s coming next—and how Decentralized AI Agents can power it.

Key Topics Covered

• The Chat→Agent shift & AI-first operations with Decentralized AI Agents
  Why the industry’s “aha” moment after ChatGPT leads to autonomous coordination: Decentralized AI Agents book meetings, research, post content, and transact. Agent-to-agent protocols cut aggregation friction and hint at a re-platforming of industries.
• ASI One and agentic LLM design for Decentralized AI Agents
  Beyond benchmarks: ASI One lets users locate specialist agents, talk to them, and delegate tasks end-to-end. Variants like ASI Fast/Extended balance speed vs. reasoning, while the “Agent” button unifies web, data, and Decentralized AI Agents.
• Decentralized AGI vs. centralized stacks
  Janet Adams argues Decentralized AI Agents and diverse neuro-symbolic frameworks (e.g., HyperON) enable punctuated leaps, broader participation, lower compute needs, and global co-evolution—positioning decentralized stacks to outlearn centralized AI.
• Deepin, Peak, and the machine economy with Decentralized AI Agents
  Peak standardizes identities, trust, and interoperability at the protocol layer so Decentralized AI Agents and machines (EV chargers, sensors, robots) can trade services/data. Energy is a flagship use case where Web2 platforms fall short.
• SMB adoption, machine-RWA, and sustainable DeFi loops
  From no-code agent onboarding for SMEs to tokenized machine assets and fee-fed ecosystem pools, Decentralized AI Agents create circular economies where more usage funds more machines—and vice versa.

Episode Highlights (Quotes)

1. “The take-off point was ChatGPT… it made everyone aware of what’s possible.” — Humayun Sheikh
2. “We’re moving from chatting with AI to doing complex things with agentic systems.” — Humayun Sheikh
3. “Decentralization will out-scale centralized AI—diverse techniques co-evolving beats one-size-fits-all.” — Janet Adams
4. “Energy generation and storage exist, but the digital market doesn’t—Deepin fills that gap.” — Leonard Dorlöchter
5. “If communities co-own robots via machine-RWA, deployment accelerates and value shares broadly.” — Till Wendler

People and Resources Mentioned

• Fetch.ai — https://fetch.ai
• Artificial Superintelligence Alliance (ASI) — https://superintelligence.io
• ASI One (agentic LLM) — https://asi1.ai
• Agentverse (launchpad) — (Referenced by guests)
• SingularityNet — https://singularitynet.io
• HyperON (framework, beta upcoming) — (Referenced by guests)
• Peak (Deepin / machine economy) — https://peak.xyz
• Lucid Labs — (Referenced by guests)
• Bosch — https://www.bosch.com
• Audi — https://www.audi.com
• Volkswagen Group — https://www.volkswagen-group.com
• Hugging Face — https://huggingface.co
• Calendly — https://calendly.com
• LinkedIn — https://www.linkedin.com
• Airbnb — https://www.airbnb.com
• Uber — https://www.uber.com
• Amazon — https://www.amazon.com
• Google — https://www.google.com
• X (Twitter) — https://x.com
• Instagram — https://www.instagram.com
• Zuber Lawler (ZuberLawler) — https://www.zuberlawler.com
• Unstoppable Domains (.podcast) — https://get.unstoppabledomains.com/podcast
• Myco — https://myco.io

About our Guests

Humayun Sheikh

Humayun Sheikh is the CEO of Fetch.ai and Chair of the Artificial Superintelligence Alliance (ASI). A founding investor in DeepMind, he is known for pioneering Decentralized AI Agents and agent-based systems that connect AI, blockchain, and real-world automation. Through ASI and ASI One, Humayun advances agentic LLMs that discover, coordinate, and transact with other agents—aimed at practical enterprise and SMB adoption across finance, biotech, materials, and more.

Contacts:
Website Link: https://asi1.ai | https://fetch.ai
Twitter Link: Hshake (as stated in interview)

Janet Adams

Janet Adams is the Co-founder of the Artificial Superintelligence Alliance and a senior leader at SingularityNet. With 25 years in banking tech, risk, and automation plus an applied AI MSc, she champions Decentralized AI Agents and neuro-symbolic approaches (e.g., HyperON) to scale beyond brittle centralized models. Janet leads global community initiatives, hackathons, and ethical frameworks to foster inclusive, sustainable decentralized AGI.

Contacts:

Website Link: https://superintelligence.io | https://singularitynet.io
Twitter Link: https://x.com/JanetAdamsAI

Leonard Dorlöchter

Leonard Dorlöchter is a Co-founder of Peak, a blockchain for the machine economy that standardizes identities, trust, and interoperability at the protocol level so Decentralized AI Agents and machines can transact peer-to-peer. With production-grade Deepin use cases (e.g., EV charging), Peak collaborates with enterprises and governments to build open, scalable smart-city infrastructure.

Contacts:

Website Link: https://peak.xyz

Till Wendler

Till Wendler, Co-founder of Peak, focuses on deploying Deepin at urban scale—bridging Decentralized AI Agents, machine-RWA, and “machine DeFi” so communities can co-own and fund robots, sensors, and energy assets. Till’s work highlights circular incentives: more usage fuels ecosystem pools that subsidize more machines, compounding adoption for the machine economy.

Contacts:
Website Link: https://peak.xyz

‍

Transcript:
Humayun Sheikh: The kind of take-off point was Chad JPT, right? So everybody realized that this is happening and it's good. I mean, when you speak to Chad JPT, when it first came out, there was nothing in the market which could actually compete with it, right? So you could see there's going to be some changes. What that did was obviously made everybody aware of what's possible.
Janet Adams: These pioneers, plus me, doing my master's degree, recognised exactly what was that this giant, ginormous shift in how the world processes data, processes information and does its jobs is coming from AI. The rest of the world went to sleep until Jack GPT came out in January 2023, whichever version it was, three, and then suddenly everyone in the world is saying, oh, we've got to use AI. But guess what? They're not really able to. And it's nowhere near fulfilling its potential in industry today.
Intro/Outro: Welcome to The Edge of Show, your gateway to the Web3 revolution. We explore the cutting edge of blockchain, cryptocurrency, NFTs, ordinals, DeFi, gaming and entertainment, plus how AI is reshaping our digital future. Join us as we bring you visionaries and disruptors pushing boundaries in this digital renaissance. This show is for the dreamers, disruptors, and doers that are pumped about where innovation meets culture. This is where the future begins.
Josh Kriger: All right. Hello, everyone. This is Josh Krieger, your co-host of The Edge of Show, live at Proof of Talk Paris 2025. I'm here with a former guest and someone I'm really excited to chat with again, Humayun, who's the CEO of Fetch.ai and the chairman of the Artificial Superintelligence Alliance. It's great to have you back, Josh.
Humayun Sheikh: Great to always be with you.
Josh Kriger: So Humayun is a tech visionary and the founding investor of DeepMind. He is also a pioneer in the future of AI through agent-based systems, decentralized intelligence, and tokenized infrastructure. And there's a lot going on under the Agentverse launchpad that you recently launched, as well as the launch of ASI One, which we'll talk about today. So I think it's worth noting that you've been a futurist in terms of seeing the potential of the intersection of AI and blockchain for quite some time. I'm curious what it was like when there seemed to be this aha moment in sort of the industry and everyone got it that there was a lot of potential here.
Humayun Sheikh: Well, I think the main kind of, the kind of takeoff point was Chad JPT, right? So everybody realized that this is happening and it's good. I mean, when you speak to Chad JPT, when it first came out, there was nothing in the market which could actually compete with it, right? So you could see there's going to be some changes. What that did was obviously made everybody aware of what's possible. And then actually start extrapolating where we're going with that. And then you realize very quickly that it's going to completely change the way we interact with AI. Because we have been interacting with AI and machine learning models for a very long time. We have prediction models which predict weather. We have prediction models which predict Amazon shopping habits. So AI has been, or machine learning at least, has been around. It's just when the user got to interact with it, that's when it makes the difference. And the same is now going to happen with how we're going to transition from just having a chat with this AI to actually doing some more complex things with this AI. And that's the next stage which I see coming. And that's, for me, that's the agentic systems that's coming. So that's the future. What also happens is with the agentic kind of networks, what's going to happen is there will be this ability to go quicker towards the AGI, which we talk about. Now, I'm just going to reserve my right on saying what AGI is because it's going to change. Interesting. The whole premise of how to evaluate if it's AGI or not, we already achieved AGI as per 20 years ago, but now we're moving into the next realm. So, redefinition of AGI is very likely. Interesting.
Josh Kriger: Tell me more about sort of how you interpret the potential of AGI at this present moment.
Humayun Sheikh: I actually, I'm a slight skeptic because I think it doesn't happen the way people say it's going to happen. There's a lot of fear mongering or there's an overhyping. But there is a actual thing which is going to happen, which is we're going to evolve. The language models are going to get better, but they're not going to crack AGI itself. But what they will enable us to do is to fast-forward the research. So what we were researching and achieving in 10 years is now possible in two days. So that itself is going to change a lot of things. I mean, I'm quite involved with, even with fetch and the models we're building, biotech models, protein, molecular protein models, building different materials, new materials, all of that is going to enable us to get there faster. So now, if you think about actually making technology which actually can be sentient, which can actually think on its own and do things on its own, is not possible today. It's not possible with LLMs because it's a very, you know, it's like retrofitting, trying to retrofit something. But because of this fast research, what you're going to do is new ways are going to come along. And I think that's what I'm really excited about.
Josh Kriger: And of course, there is that sort of doomsday theory as well on the proliferation of agents that are already taking over half of the Internet traffic. And, you know, this this idea of sort of, you know, you tell an agent generate fifty dollars of wealth and it figures out nefarious ways of doing that. You know, the complexities of sort of, you know, autonomous agents earning money or transacting and jurisdictional complexities there. But it seems like there's more of a positive color in terms of the future potential of AI than there was six months ago, perhaps. What are your thoughts on that balance and sort of the ethical implications for society?
Humayun Sheikh: There's a lot to unpack. What do I think? Where are we going? And why has the sentiment changed? People have realized that it's not going to take over everything. We were over optimistic. We have this problem in crypto. We have this problem in AI. We've done it since 1950s. We have the summer of AI and then we have the winter. We have the same cycles in crypto. We are optimist, and then we become over-optimistic, and then we become pessimistic, and we become over-pessimistic. So that's a normal cycle, right? But there is a reality to this whole thing, which is, when web came along, there was a lot of chatter about that as well. We stabilized, we started doing e-commerce, we buy things online. When people used to say, well, this will never happen, So what is going to happen is, first thing that's going to happen in my opinion is, we're going to change the way we do things. We will become AI first. We were becoming web first. So when the web came, we became web first because everybody, there were shops, online shops and everything. And they were built based on this HTTP architecture. We built this e-commerce architecture, and then we deployed it, and then we started using it. Made life easier. Amazon was born. Google was born. All of these things were born out of that. We're now going to see the same thing happen in AI, which would be AI first. Now, so if I take the same parallel, so let's take the technology, the infrastructure. Now we have this technology, the infrastructure and the user interface. User interface was quite important because otherwise how would you interact with it, right? So now, let's take the same parallel and move it to AI. User interface is kind of getting very resolved. Why? Because it's language. It's natural language. You're going to speak in any language. You're going to look at things and it will deliver you things. So that's the user interface covered. I'm just creating this parallel, the website parallel, the web parallel to AI parallel. We now have this engine which is thinking, the large language model, because it's actually reasoning is thinking. In Webb's case, you were just, you were doing that thinking, and you were then going and interacting with these things. Now, something else is doing the thinking, and you now need to interact with something else, right? So businesses, person, everything. How are you going to, how is this thing, this AI, going to interact with them initially is using agents. Right? So you have an agent, I have an agent, your business has an agent, my business has an agent, everybody has an agent. Yeah, and essentially they become members of the team. It becomes members of the team. And then, so now you come along, you go to your agent and say, hey, find me Humayun's diary and book me an appointment with him. So your agent will go along and they will try and connect with my agent. and it will then look at my diary and look at your diary and it will arrange that. We do that through Calendly or Google Mail or whatever we now but that's going to change because finding a simple thing like finding calendar invites to each other where the times can match is a nightmare still. Yeah, it takes a lot of coordination. A lot of coordination. Take it out. That's AI first. So what that enables is that your lives are now becoming more run by AI. So that's where the transition is going to come. We were running by web. Now we're going to be run via AI. I may call it AI, AI tools, whatever you want to call it. And then what would happen is you're going to generate huge amounts of data just by doing the interactions. And then this AI is going to learn more from that. Now, you could do a runaway, like a train, where you can say, well, everybody's generating so much AI data, which it's going to absorb, and it's going to learn so much, and then it's going to be in control. Okay, maybe that could happen. But how can you protect it is just making sure that you put it in your control. That's really where we're going to. And I think that's where the regulation comes in.
Josh Kriger: Very cool. Well, thanks for sharing your insights there. And I think that's a great transition to talk about ASI 1. So you rolled it out a few months ago and it really puts LLMs in the hands of users. Tell us a little bit about what makes ASI 1 different than some of the other AI projects that have been on the streets or are coming out soon and what gets you excited about it.
Humayun Sheikh: Right. So as an LLM, it does everything any other LLM would do. It's up there with OpenAI. We've done all the benchmarkings and it's kind of competing with all of that. That's not the fun part. That's not the exciting part. The exciting part is what that results in is the communication between agents. So you can actually go today, you can actually try, you can go in and you can ask it to find you an agent which can do something. And then it will find you that agent, you will actually say to it, oh, go to this agent and connect with it and ask the agent what it can do. So now you're speaking to somebody else's agent. At the moment, all the LLMs are doing is you're speaking to the LLM in the middle, and everything is communicated through them. What we have done is we've taken that part as well, so you can actually say, hey, do some research somewhere and say, find me a company which can do best supply chain in France or Paris, right? For example. And it will talk to other LLMs. It will talk to other LLMs. Other agents. It will talk to other agents. It will find you the agent and it will say, hey, you guys are now connected. What do you want to ask this agent? That agent doesn't have to connect with a human straight away. It can say, hey, I can give you all the information you need. I can give you all the information that's on my website. or on my database.
Josh Kriger: So it sort of skips a step that has been sort of an inefficiency in our world. Like, you know, I want to book a trip to Istanbul. It will find the various travel agencies that can get me there.
Humayun Sheikh: Actually, probably not, because it shouldn't do that. Because the whole inefficiency resulted in aggregators. So do you see the point here? The point is travel agents are not needed because travel agents were removing that friction between a travel company like a hotel or a flight and the consumer because you can't access each other. So that's why the travel agents came. That's why the aggregators came. So this changes the whole platform concept throughout many industries. Many industries because what will happen is I mean all the big companies today are aggregators. Airbnb, Uber, Amazon, Google. All of them are aggregators. They're aggregating something. But you don't need that. So you and I can connect. LinkedIn, aggregator. You submit your details, I submit my details, we find each other. No need. I say to my agent, go find me somebody who is doing interviews and is called Josh. It finds you. It connects with you directly. You don't need to register anywhere. As long as you have an agent, I have an agent that exists, we can find each other. Yeah, this is quite game changing for... Which if you listen to anything I've said, like for the last seven, eight years, that has been our mission. Our mission is with agentic systems, you don't need this inefficiency. And if you remove the inefficiency, the game changes. And that's really what's happening in AI world. If you become AI ready first, AI first approach, you get rid of a lot of inefficiency.
Josh Kriger: Noted. So on that topic, you've had some recent benchmarks with ASI Mini, which are quite impressive. And, you know, we were just talking about how decentralized AI can compete with the centralized giants. Tell us a little bit about ASI Mini and what the recent sort of indicators are in terms of that potential.
Humayun Sheikh: Yeah, ASI, I mean, ASI Mini is probably now slightly old news. I mean, we call the LLM interface is called ASI One. Now we have multiple models within that. So ASI Mini is one of them. We have ASI Fast, which is, which responds a lot quicker, less complex in terms of reasoning. Then you have ASI Extended, which does a lot of reasoning. So if it's a complex task, you can give it that, and it will think a lot more. It'll go through all the reasonings. But the exciting bit, as I said, is you can actually click an agent button, and it will then correlate what it can find on the web, what it has in its database, and what the agents are available, and then it combines and gives you a final result. and it tells you, for example, just giving you an example, you say, hey, I want to create a tweet and I want to post it, right? I want you to create a tweet for me. You can ask it to create a tweet. You can say, hey, is there an agent which can actually give me some factual information about something? It'll go and find the agent, it'll get you the right information, and you can say to your agent, hey, now go and post it. All from one place. You don't need to do integrations. You just type away and you say to it, and it says, okay, submit your credentials for Twitter, I'll post a tweet. That's the kind of thing we're talking about. So that's what I'm more excited about. The benchmarking is okay. We are at par with all of them.
Josh Kriger: But again, because everyone, all the LLMs are talking to each other and evolving in parallel, it seems like that sort of element of the LLM world is going to continue. It's just going to be more and more similarity relative to differences, right, between the big ones.
Humayun Sheikh: Yes and no, because our architecture is slightly different. Although it does the same job, it's a different architecture.
Josh Kriger: I guess, can one LM leapfrog the others at this point, or will it be marginal?
Humayun Sheikh: It depends on what. It depends on, are you just talking about knowledge base and text? Or are you talking about doing new things? So yes, it can leapfrog because it can learn, if it has more interactions, it can learn better, but in one space. But on the other side, if it's more tuned to doing some other things, like for example, ours is agentic and there's no other agentic LLM available. So of course, we're going to be better at that because we are doing it. So, but learning from each other is quite an important part. I mean, we're releasing, by the time you kind of release this, there'll be another feature where you can actually, your agent could be using one LLM, my agent could be using another LLM. I can put them together, I can say, go and argue how is Trump's government doing in the world, right? And they can take separate sides and they can argue and make some good reasoning and try and convince, which then gets fed back into the LLM. which it learns from. So which means that the more these kind of interactions happen, it will actually self-learn, it generates the data, so it accelerates that part. So yes, they can leapfrog, but I think we need to see some new technology kind of coming in rather than just building on the same LLM. So there is going to be breakthroughs which I expect over the next few years.
Josh Kriger: Well, and on the new features and sort of partnership side of what you're doing, there's some enterprise relationships that are, I'm sure, have been forming for quite some time and some that already exist. How are you looking at sort of leveraging ASI for enterprises moving forward?
Humayun Sheikh: So that's quite interesting, because I personally took an approach on Fed's side because of my history in the past with DeepMind. My approach is not to take enterprise, because I don't want large enterprise. I don't want them as our customer. There is a reason for that. Although we worked with them, we also kind of go through the whole process with them. We find that it's very difficult, especially in where these hyperscalers are kind of available and these hyperscalers are very hungry. It's better to go for a niche, smaller market. So if you think about small and medium businesses, they are nowhere on AI scale right now.
Josh Kriger: And they actually are being stifled. And I think, you know, when we look at economic balance in our society, as a small business owner, I would like to see small businesses thrive again.
Humayun Sheikh: Yes. And so we took that approach. We took the approach of individuals in small and medium enterprise. And the reason why we took that is these guys don't have the IT skills. They don't have the AI, the resource. The resources and human resources and just human resource management is much more challenging than it used to be. It's very different. So we're providing solutions for them. So we're saying out-of-the-box solution. You don't need to code for an agent. You don't need to code to do any of this. You take your legacy system. If you have a website, you should have an agent. And it should take you two clicks to get there. And that's what we're providing. So you can you can get your agent up and running within five minutes with no coding experience. And once you do that, then you start the journey of learning and we kind of bringing all of that to you. But that is not to say that we're not working with enterprise because what smaller and medium enterprises realized is that the solutions that these hyperscalers are providing are not niche enough. They don't give you niche opportunities. So, for example, we're working with companies on NASDAQ at the moment, smaller size, but they are looking to bring the technology and bring AI into their business, which they have very little at the moment. But if we can accelerate that propagation and build solutions for them, then you suddenly start seeing these companies also getting into the AI game. Because otherwise, if you don't use AI, you're going to be left behind, right? So you have to be in the game. And they realize... That's to be part of your culture. That's the point. Of course. And that's what they're realizing, that they will not survive if they don't do that. So we've seen some really interesting uptake in quite a lot of these companies, which are not too small, but they're small to medium. and they're bringing AI in different ways. What is quite interesting, and I think in the context of crypto as well, crypto is always lacking a use case, right? Crypto is a technical... It's finding a solution without a problem. Without a problem. So what my approach is, forget about crypto, let's see if we can build a solution which people want to use first. And that's what ASI 1 did. That's what Agentverse did. So we bring in the agentic solution, we bring in the LLM to the masses, and then, without telling them that it's crypto, without them having any interaction with crypto, but underneath, it's decentralized, it's crypto, you have all of those things. But they don't need to see that. They don't need to, we can't just say, oh, well, we are an LLM, and what's your uniqueness? Well, we are decentralized. No, that's not a good thing to say. What we need to say is we have these differences. Oh, by the way, it is decentralized. You don't use it because it's decentralized. You use it because you like the features in it. And that's what we've achieved. And what we're seeing right now is like the uptake is like exponential right now because people are actually finding the features useful. They're actually using it. But underneath, you're generating transactions, you're doing all of the crypto stuff. But what is also interesting is that for them to build these big solutions, enterprise will have to have the FET token, right? So we're bringing a non-crypto market to crypto. So that's, and you will see that, I think there's a couple of announcements going out in the next day or so.
Josh Kriger: Well, congrats on all the success. Before we end, I know you're an avid user of AI. I think we all get overwhelmed with the possibilities of what we can do with AI. What are some of the fun and novel ways you use AI just in your day-to-day life that makes life better?
Humayun Sheikh: So when I'm trying to research something, I use I mean, I use all of the LLMs, but of course I have a predilection towards ASI-1. So I use it, I give it a character every time I do that. So for example, I could ask it to act like a degen, crypto degen, or I could ask it to be a physicist or something. I ask it to be my general counsel. I quite enjoy doing that because then you can see all the emojis and all the sarcasm come through, but you're still learning what you need to do. So I find that element quite interesting. When you get using it and when you start speaking to other businesses and when you start speaking to their agents to do something, people have different ways of dealing with it. So you can say, your agent, when goes to the world, should represent your personality and that's my personality I want to represent. And then when you interact with them, you see the personality come through, which kind of humanizes the whole thing a little bit.
Josh Kriger: What was one of the most novel things you personally have done with AI the last 36 hours?
Humayun Sheikh: Oh, that's interesting. Well, I've written a PR. I've done a kind of an agentic code, which took me only half an hour, to build an agent which could do something very interesting, do the research. So that's kind of the couple of things which I've been using it for. But I think Actually, some of the use cases which have come out of it are quite interesting and we're implementing them. So not perhaps on the 36 hours, but maybe last week. I mean, I asked the ASI one to give me suggestions of what kind of agents should I be building and what should be going on. And it actually has given me, I mean, seven out of 10, we are going to implement them.
Josh Kriger: Can you please speak to one of those agents? I think he'll already be out by the time this comes.
Humayun Sheikh: So we have an agent which is doing the deep financial research on tokens. Oh, that's great. Which is that you don't need to go to this platform, this platform. It kind of collates everything and actually generates a report for you. So you can tell you what the wallets are holding, which network they're holding it on, what's the inflows, outflows. It kind of gives you a sentiment analysis. It picks up everything from X, from Insta, from whatever. It kind of collates it and puts it together.
Josh Kriger: Yeah, I mean, I could think of a lot of products that you're going to run for their money with that type of technology. Really exciting stuff. Really appreciate your time and your insights today. Well, how was it? Pleasure to hear this conversation. Where can folks go to learn more about Fetch and ASI and all the things you're doing? And I'm sure there's going to be some curiosity and folks will sort of test out.
Humayun Sheikh: Yeah, so if you want to test out the LLM, try the agentic stuff. It's asi1.ai. Pretty short and sharp, but I would highly recommend you try it out. And if you want to know more about it, just ask asi1. It'll tell you everything about Fetch. It will even write you the code to do an agent. It will send you to the right place. It will give you business opportunities to build. And we talked about the launchpad. So what we're planning to do is that if you have an agent that you think other people are going to use, you can raise money around it. But we're not looking at the meme economy. We're looking at true application. So we're looking at people building useful agents, which industry, people, everybody can use. And if you have those ideas, I would highly recommend, again, going to the launchpad, and launching your agent so that people can see it and actually you can raise money around it and you can actually build machine learning models and you can connect them with agents and you can launch them and you can raise funds. I mean we have just onboarded roughly 200,000 odd hugging face machine learning models onto our system. You can actually go on ASI One, you can query them You can get the output from them. You can do research on it. So it's one place to find all.
Josh Kriger: Wow. And if folks want to follow you and some of your thought leadership on X, what's your handle?
Humayun Sheikh: Oh, that's interesting. It's Hshake. I'm sure you can find me. All right. Because if you type my name, I don't know. I don't remember my handle.
Josh Kriger: Yeah. That's what you're agents for, right? Yeah, exactly. Thanks for your time today. Thank you. Appreciate it. Let's take a pause to shout out one of our favorite partners. For tech innovators facing legal challenges, Zuba Lawler is your go-to law firm. They focus on understanding your technology and business model before addressing legal requirements. Specializing in blockchain, AI, VR, AR, quantum computing, and more, ZuberLawler offers expert guidance in capital raising, IP transactions, M&A, litigation, and compliance. Visit ZuberLawler.com, that's Z-U-B-E-R-L-A-W-L-E-R.com for cutting edge legal solutions. Hi, everyone, and welcome back to The Edge of Show, live in Paris at Proof of Talk. It is a bustling here, and I'm so excited to have a former guest back on the show. I think we last got together in Dubai. This is the COO of Singularity Net, Janet Adams. Great to have you back.
Janet Adams: Josh, thank you so much for bringing me back on your show. Hi to all your listeners. We're in for a treat today, I would say.
Josh Kriger: I would just add, I'm no longer just the COO of... Also the co-founder of the Artificial Superintelligence Alliance. We had Humayun on the show as well, and he talked a lot about what's happening in sort of the ASI sort of area, which is really exciting. So, I guess, like, what's your take on how AI is going to impact the industry as a whole? sort of, you know, how we got to this moment where it was like pioneers like Singularity Nut and Fetch that had these ambitions and all of a sudden everyone woke up.
Janet Adams: Everyone woke up super late compared to us. So in 2017, I was working in banking. My job was in global banking technology and risk in corporate and investment banking for 25 years. And I read a Harvard Business Review article in 2017, which It was about AI and industry and it concluded that AI won't replace managers anytime soon, but managers who use AI will replace managers who don't use AI.
Josh Kriger: I love that.
Janet Adams: Isn't it great? And I just got super inspired and decided to do a master's degree in AI, my second master's degree. And so I started my degree and my master's degree and I couldn't believe I got in at the University of Essex to the UK's only applied AI MSc. Congratulations. Thank you. At the time I was like, surely everyone in the world, every single executive wants to do an AI MSc. Next year, it's going to be completely overrun with executives. And the funny thing was, next year, there were no executives doing the AI MSA. I'm like, where is everybody? Why is everyone so asleep? Why does nobody see that this AI revolution is coming and it's going to be big? The rest of the world went to sleep, more or less, for another six years, I would say. Ben Goertzel founded SingularityNet, of course, in 2017. Trent McConaughey and Bruce Pond founded Ocean Protocol in 2018. Fetch, Humayun Shaikh, founded Fetch.ai in 2019. And these pioneers, plus me, doing my master's degree, recognized exactly what was that this giant, ginormous shift in how the world processes data, processes information, and does its jobs is coming from AI. The rest of the world went to sleep until ChatGPT came out in January 2023, whichever version was three. And then suddenly everyone in the world is saying, oh, we've got to use AI. But guess what? They're not really able to. And it's nowhere near fulfilling its potential in industry today for a number of different reasons. Now, I have a long career of robotic process automation in financial services, of various forms of workload automation and process redesign. We've always had this dream that you get efficiency out of automation and intelligent automation, but it almost never works. The ROI is not there, Josh, in industry, and it's still not there today because people are using narrow, limited, large language models, which are based on deep neural networks. For all of their AI implementations in industry, they are opaque, they lack creativity, they're brittle, and they can't cope with any data instances which are outside of their training data set.
Josh Kriger: Is that a derivative of the fact that these are centralized AI models? And does decentralized AI just by default, you know, offer those things? Or if not, what's the heavy lifting that is required to make decentralized AI more competitive or preferable to centralized AI?
Janet Adams: Okay, so my first degree was in geology and I studied evolutionary sciences. And there's a type of evolution called phyletic gradualism, where things develop like slowly, like the frog that eats the most worms has the strongest babies, all this type of stuff. But then there's a new theory that was introduced while I was at university of punctuated equilibrium, which is that competition and evolution doesn't happen in gradual steps. It happens in one, it happens in giant bursts where there's a huge dramatic change. That's what's coming now. So it's not that decentralization is going to start to compete with centralization, it's that decentralization is going to And the reason it's going to take over in technology and in AI is because in a decentralized development scenario, we can scale. We're releasing our HyperON AGI framework in beta later this year from SingularityNet, and that enables every diverse kind of AI to interact with each other on a hypergraph.
Josh Kriger: So clearly you have a lot of conviction on the power of decentralized AI, but what if it doesn't? What are the sort of potential roadblocks here that could change the trajectory or slow it down?
Janet Adams: I can't think of one. Because big tech, like you say, big tech is focused on one kind of AI. It's deep neural networks. We're not. We're beating that. We're using evolutionary programming. We're using neuro-symbolic techniques, more powerful AI techniques. And in decentralized AI, we have millions of developers coming out of university every year. Those developers can go straight on our agent-first today. They can go straight on our ASI Create platform and build agents. They can go straight on to our hyper on beta when it comes out in beta, and they can make better AI, and that AI can co-evolve in diversity all around the world and get better and better and faster and faster. And this is why this is going to be the single first example in humanity where decentralization is going to beat centralization. It's going to be a winner-takes-all. Well, winner-takes-all, winner being decentralization and decentralized AGI technologies.
Josh Kriger: So it feels like the subtext here is that the quality of the product will market itself, that it's product first. Is there any concern in your mind that these big companies with large earnings have unlimited marketing dollars and that those dollars could sort of, you know, push them up the SEO rankings and into the minds of folks where it would potentially cause some challenges with adoption of decentralized AI.
Janet Adams: You raise an excellent point, and you asked me that question.
Josh Kriger: It doesn't sound like it's about product, yeah.
Janet Adams: What could stop us? Big tech has such a huge amount of money, such a huge amount of compute, but we have an artificial superintelligence alliance, we have scalable decentralized, we have the biggest AI high-end deep-in network in Web3, so we can scale. We've got the compute power. We've got the data, we've got the blockchain layer, we've got the agents, we've got the AGI, we've got the neural symbolic techniques. But yes, it's possible that somehow or another, big tech could squash us, or it's possible that one of them might actually come up with an AGI breakthrough rather than us first. I personally think it's unlikely because they're basing all of their AGI plans on deep neural networks. And it's just, it's too static. It's too brittle. It's too rigid. It's too, it's too undiverse or whatever the word for undiverse is. Whereas we have the entire diversity of the world and the entire, because we use neural networks as well. But we're also more environmentally sustainable. Our techniques use need less compute power.
Josh Kriger: I mean, we as Singularity or the... We as ASI. ASI.
Janet Adams: The stack. It's the whole stack.
Josh Kriger: It's the whole stack now.
Janet Adams: That's incomparable in either Web 2 or Web 3.
Josh Kriger: Pretty amazing, and you've put some real context into decentralized AI that we haven't discussed this, so I really appreciate that. There's one more topic I just want to wrap up with, which is sort of inclusion in this new economy and diversity. What is your perspective on how SingularityNet can sort of advocate for this type of outcome?
Janet Adams: Beautiful last question of the day. Thank you. We are wholly committed to diversity. We have the most diverse leadership team in technology. We are gathering ethics from all around the world. We're doing hackathons in India. We're doing hackathons in Africa. Our goal is that when the AGI is making decisions, that it's making those on behalf of the most of the many, not for the needs of the few. So we're utterly committed to diversity, diversity of developers, diversity of community, diversity of algorithmic approaches, diversity of ways of thinking, and all forms of great flourishing in this age of AI.
Josh Kriger: Great to catch up, Janet. We'll have to go deeper at some point soon. If people want to learn more about SingularityNet and follow you, where should they go?
Janet Adams: Please go to the Artificial Superintelligence Alliance website, superintelligence.io. Please follow us on Telegram. I'm at JanetAdamsAI on Twitter and also on LinkedIn, so come find us.
Josh Kriger: Great to catch up.
Janet Adams: Thank you, Josh. Fantastic to speak with you and your team today.
Ad: I learned a lot. Thank you.
Janet Adams: Thank you.
Ad: Hey listeners, are you ready to amplify your voice in a digital world? Introducing .podcast domains, your gateway to the future of online identity. Whether you're a creator, a company, or just someone with big ideas, .podcast is for you. Own your unique digital identity, connect securely, and explore hundreds of apps with no renewal fees. Don't wait for someone else to claim your name. Secure your .podcast domain today and be part of the next generation of content creation. Visit get.unstoppabledomains.com backslash podcast to get started. Again, that's get.unstoppabledomains.com backslash podcast. Your voice, your rules, your identity.
Josh Kriger: Hi everyone, welcome to the Edge of show live at Proof of Talk Paris 2025, where there's a lot of different aspects of the industry coming together. And one of those is the future of Deepin. So I'm really excited to be here with the co-founders of Peak, Leonard and Till. Great to have you on the show. Great to be here. Likewise, great to be here. So you all have a really interesting project that folks should know about. And we'll get into some of those details. But let's start, Leonard, just sort of define this idea of this machine economy that we're in today. And how is Peak enabling it from concept to reality?
Leonard Dorlöchter : Absolutely. Yeah, so the journey goes back actually all the way till 2017, where we for the first time thought about the machine economy. That's right when I got into the industry. It's been like eight years now when we thought about, okay, at some point machines will be able to pay each other autonomously. For example, a charging station and autonomous electric vehicle. It's probably the easiest example to imagine. And we thought, okay, this is going to happen, and we had the feeling it will happen with blockchain. And then over a couple of years, we first of all started working with big enterprises, such as Audi and the Volkswagen Group, because we wanted to figure out, are there real problems to be solved? Are there real business cases to be built? And they actually said the charging infrastructure is super fragmented. You have all those closed centralized platforms and users need to sign up with like 10, 12 different charging platforms in Germany alone. And they also thought there must be a way how a private charging station can be made accessible to the public because an Audi manager always gave his father-in-law five euros after charging at his home because the car couldn't pay the charging station peer-to-peer. And they thought maybe, okay, blockchain could be the solution to solve that fragmentation. And this was just one use case where we realized, okay, that problem exists in charging. But if you think about a smart city of the future, everything, every machine, every device, every human needs to be able to interact with anything and everyone around them. In a safe and secure, non-chaotic way. Exactly. And in a way where they can trust. Because when you do a peer-to-peer transaction, you need to be able to trust. And that's why you have centralized platforms, trusted third parties, who take care of transactions. And if we want to create an efficient peer-to-peer machine economy, we need an infrastructure that creates trust at the protocol level. And this is why we started Peak as an own blockchain, because we said we need to build those trust functionalities in the protocol of the blockchain itself. And then everyone building on it has that foundational root of trust, can build applications, peer-to-peer transactions, and this is kind of the evolution.
Josh Kriger: Very cool. And what was the first actual use case built on Peak? And when was that?
Leonard Dorlöchter : Yeah, so it was actually the charging use case where we then realized after a while, okay... Approximately what time? And so that started working with Audi in 2019 to 2021. 2021, we started building Peak as we know it today. And then in 2022, the first decentralized charging case. was built. It's also part of the European initiative where other big companies like Bosch are getting involved and so on. And that was the first use case.
Josh Kriger: So I just did a little AI research while we're talking and this was before Helium.
Leonard Dorlöchter : Yes, like the first case, the first Deepin case that we thought about. We didn't call it Deepin and it was in the enterprise context. But yeah, the concept we started wrapping our head around in 2019 already.
Josh Kriger: Cool, and now how many different applications are built on Peak?
None: 60.
Josh Kriger: Okay, so a lot's going on. And Till, you recently wrapped up the Machine Economy Days in the UAE, a bustling sort of innovation hub. We do an event in Riyadh and Dubai are sort of competing to be the smart cities of the future. And as a bystander and participant in those economies, it's really fun to to experience like this innovation in real time happening and all the exciting sort of technology being built with the government. What were sort of some of the big takeaways from that experience and how has that sort of influenced your attraction globally?
Till Wendler : Yeah, I mean, when it comes to those events, I think everyone who has seen what we've been posting on social media, that those were not the average, let's say, token 2049 crypto events. So those were, they were governmental officials, some of the representatives of the largest companies in the UAE. Yeah, a lot of highly influential people. And, yeah, it really goes, as we know the UAE, it's always looking to be one of the most advanced places in the world and I think it's definitely on track, if not even, yeah, is already one of the most or the most advanced. And yeah, when we're looking now at the cities of the future, like the smart cities of the future and also the life quality for the people living in those cities, then the Deepin model can play a significant role for those cities. Because it's much more than just deploying hardware infrastructure. It really goes into engaging citizens in building the city of the future. And this is something that the UAE sees within the Deepin model. And specifically, that's why we also push certain initiatives there. I don't think I can share too much yet. I hope I will later today.
Josh Kriger: Yeah, were there any like particular use cases that came out of that experience that, you know, just sort of got you excited? Because, I mean, you've seen a lot of different use cases and you're thinking about Deepin creatively every day. I'm just curious if there are any sort of advancements made that really sparked you?
Till Wendler : Indeed, yeah, some of the big cases or some of the big initiatives that we see is when we think about physical machines, right, and when we look at the broader crypto space, then a lot of the things are happening in digital nirvana, if you want, so right, but when it comes to physical machines, they need to be somewhere, they need to be at a physical location, and that's why it needs some sort of regulatory support. Specifically, when it comes to larger asset machines, when it also comes to the financing of those machines, for example, looking at machine RWA, definitely one of the big topics that I think we can push there very much. Again, I'm not allowed to say too much yet, but yeah, definitely one of the bigger topics. But then, yeah, like really looking at a lot of those sensor deep-ins, for example, data deep-ins where, yeah, various kinds of data is being generated, be it noise data, be it air pollution data or whatsoever. In the end, it also needs a demand side. And yeah, that is something we really see that they are very interested in that data, in engaging the citizens. And yeah, that's what we're really looking for now, deploying some of those data deep ins, really go deeper in machine RWA, those kinds of topics.
Josh Kriger: Right on. And Leonard, you mentioned earlier over 60 projects deployed. I think that makes you the largest, is that fair to say, in terms of platforms for Deepin? By number, yeah. And what's your overall sense in terms of the traction of Deepin? I feel like There's definitely some criticisms out there that Deepin is a solution without a problem, and I'm sure you would argue differently, but where do you see sort of the evolution of Deepin at this moment, and what is it about Peak that's sort of attracting all these projects to your platform?
Leonard Dorlöchter : Absolutely. Maybe let me answer the question with one concrete example. Because if we look at the energy market, for example in Germany, there are enough renewable energy assets already to power the entire country. But the problem is, there is no digital infrastructure to trade the energy efficiently. And there's even enough storage. You have enough electric vehicles who could store a lot of energy and release it back into the grid when it's needed. So technically storage is there, energy generation is there, but the platform is missing. And my brother actually, he worked, he did an internship and in that internship he realized Why we do what we do? Because they discussed having to build a centralized platform to manage all of this, but also realized it's almost impossible to build, let's say, a Euro wide platform that manages all of the energy. So you need a blockchain, you need energy deepens, which are building up, which are connecting all of those assets to create a super efficient energy market. And this is just one… What if the infrastructure just doesn't exist? It doesn't exist. And like you need to have many microtransactions. You need to be able to verify how much data went in and out. And you need a unified system that's open where this is happening in real time and global, right? Or at least in one continent, right? Energy is just one area where there's a very clear problem that cannot be solved with Web 2.0 technologies or it's very difficult and very inefficient and where Deepin is creating a grassroots movement where every energy asset at some point is connected and is able to share data, consume and sell goods and services.
Josh Kriger: So you're now dealing with like cross-border transactions and sort of a mix of regulations. So are you all thinking about sort of the regulatory environment and how that's going to affect the industry?
Leonard Dorlöchter : Absolutely, yes. So that's also another reason why we work with governments, want to build regulatory partnerships and frameworks, and we do pay close attention. And actually, a lot of the regulation in areas of the world is benefiting deeply. In the EU, for example, European Union, there's a lot coming up which is going to benefit decentralization in terms of how energy markets are being managed, how identities are being managed. and maybe to answer your question on why projects choose P. Yeah, what is it about your platform that no one else is offering right now? So it's like going back to those protocol native standards such as machine identities, where we don't think one individual deepen, we think about how do we enable all deepens to seamlessly work together. So if two energy deepens build on peak, they are automatically interoperable if they cover slightly different parts of the problem. And we built this standardized interoperability on a technical level, which no other chain does. On other chains it's fragmented, it's on the smart contract level. We create a trust system also on the protocol level, which, because it's impossible to verify truth on scale in those systems. It won't work, but you can hold actors accountable. And this is also what we're building and providing. And then additionally, the go-to-market, like we work with those big companies like Bosch, we work with governments, we help our projects go to market, establish new regulation, and build up sustainable revenues. Because a lot of deep-ends are building in markets that don't exist yet and it needs a lot of time and support. NVIDIA did this. They always built markets in markets that then existed 10 years later. And quite a few deep-ends create value on data that doesn't have a market yet and they need support in creating it. And yeah, we're creating that meta ecosystem basically for deep-end and the machine economy.
Josh Kriger: Thank you for that helpful context. These partnerships with large players seem critical to the advancement of the industry. Till, you guys recently announced a partnership with Lucid Labs. Congrats on that. What does that unlock for Peak and for your community?
Till Wendler : Yeah, I mean, Lucid Labs or the VEOs that we do with Lucid, they're part of a larger initiative that's happening on Peak. Actually, it's one of the, if you want, one of the three core pillars, what we believe are the journey also to a completely decentralized, full-scale machine economy, which is basically the infrastructure that we need to provide, the access to intelligence for those machines, but then also financial ecosystem that need to exist, what we like to refer to as machine DeFi. whereby you can think about more traditional DeFi protocols coming onto peak and deploying solutions on peak, be it a DEX, be it borrowing and lending, but also be it innovative products like the VEOs that Lucid provides, for example. But they do it in a slightly different way. So for example, if we're looking at what I described before, right, the tokenization, we're speaking about machine RWA, in the end, for example, having a share, let's say in a in a tokenized service robot, for example, that provides any kind of service or resource to people, being able to use that share then, for example, that tokenized share in a borrowing and lending system will, we believe, have a complete different impact on DeFi in general. Because out of the sudden you have a real world revenue generating asset. Let's say it's RoboCafe for example.
Josh Kriger: Yeah, I mean, that makes a lot of sense. I've sort of had a lot of recent encounters with humanoids myself. We had Rhea on the Edge of show. She's a humanoid specializing in mental wellness support. And, you know, even in the pilot phase, it was quoted on the show that, you know, she'll cost a little over $100,000. But imagine if a community could all have a stake in her being out in the world. I think there's a scalability factor deployment acceleration that's possible with different types of humanoids with different specialties that can improve society.
Till Wendler : Correct. Yeah. And we, I mean, we see that happening more and more, right? Be it with more simple robots, as I said, right? For example, service robots like RoboCafé that's replacing the traditional barista job or be it a delivery bot or delivery drone that's replacing the delivery jobs. Autonomous vehicles, obviously. And then there are two ways to go about it, right? Either they are owned by large corporations, that basically deploy them and own them also completely, or we can own them directly and all having, yeah, a share.
Josh Kriger: Yeah, I mean, I think about grocery stores, like, yeah, the self-checkout is great, but no thanks. I would rather be, you know, talking on the phone or doing some productive work, like, let's get some robots in these grocery stores, like, you know, helping us check out of the grocery store, right?
Till Wendler : Yeah, and I think for now, it's also some sort of natural defense mechanism that we don't necessarily allow that really to flourish. Obviously, there's also technical developments that still need to happen. But if we understand that we can have a share in this and we can actually profit by the output of those machines, by those robots, and we can profit, all of us can profit by it, and also the average person out there can have a share in those robots and systems and autonomous machines, then that's very powerful. And that's why it's very interesting for a lot of DeFi protocols and DeFi projects now to come to Peak, to pull on Peak, because they see that, they not only see that machine RWA or tokenized machines can be extremely powerful and an actual real revenue generating asset that can be used in those protocols.
Josh Kriger: Yeah, just on that topic, I mean, we talked about what you're doing in the UAE and how you're sort of supporting the ecosystem generally. Can you be more specific, Till, and maybe talk to us about, you know, anything from launchpads to grants, you know, funding strategies that you're thinking about to push the industry forward?
Till Wendler : Yeah, it's very diverse. So it's, I mean, for looking, for example, looking at that creative ways, right? So for example, looking at Lucid, that's, I think, a really good example, how we also want to, you know, engage the community a little bit more into our funding strategies and so on, giving them access. through those VEOs, then using those funds, for example, deploy it into different DeFi protocols, like the proceeds from those raises, deploy them in the different DeFi protocols that are running on Peak.
Josh Kriger: And then... And encouraging long-term sustainable economies, right?
Till Wendler : Correct. There's, for example, part of our system or part of our network are those we refer to them as like, yeah, like network native or systematic system pools. So pools that are being filled by the transaction fees, part of the transaction fees of the network. But now also those DeFi protocols plugging a part of their fees into those pools, so the more trading volume on those different projects, DEXs, launchpad raises, borrowing and lending systems, technically the more money goes into these system pools. And they are then being used, for example, for machine subsidization. So in the future we'll have also governance systems around like projects applying them for this machine subsidization, being able to subsidize parts of their machines so that we really create also Yeah, a very natural loop in the in the ecosystem that the more volume of more trading volume, the more transaction volume in the system also created by machines results in more machines being able to be deployed. And yeah, really trying to create a machine economy that also becomes very autonomous and sustainable into funding also itself. And of course, we have traditional ways, if you want so, in terms of working very closely with some of the VCs that are deploying capital. I think they deployed around 15 million in the last year, 10, 12 months, into different projects, building on peaks. We really see quite some movement there. But yeah, working a lot also on those natural flows in the system.
Josh Kriger: Thanks for sharing all your insights on where the industry is at today, gentlemen. And just to put a pin in it, let's fast forward two years from now. Where do you think Leonard will be with sort of the evolution of smart cities and sort of dpen in your sort of most optimistic view and a more realistic view? And feel free to add to the conversation.
Leonard Dorlöchter : Two years, it's two years, not a long time. But at the same time, right now, the advancements in AI are crazy and robotics as well. So I do think in two years, we have significant, deepened infrastructure deployed in smart cities. at least in those embracing the technology. So I'm very forward looking countries and smart cities. I think there will be a few very clear product market fits where deep ins have reached a lot of scale revenue and where smart cities clearly see, okay, this is a better way to deploy infrastructure, involve our citizen and create a better smart city as well because there will be more data, there will be more engagement, And, yeah, people will be having a higher quality of life. That being said, though, I do think it's going to take five to 10 years to really have a full scale, deep and nice smart city where we see also machine economy, like where we see a lot of economic exchange across different machines and devices, like so that the mainstream becomes more aware of it. It will happen in the background, but that this is, yeah, in a more medium view, I'd say. So that was the media view actually in two years. Now the optimistic view is that few countries which are really embracing deepen. For example, in the US, there's a lot happening, right? The Emirates. Who are Emirates? Emirates as well, right? Politicians and cities actually say, okay, why don't we can build with deep ends, we're going to build with deep ends, we're actually going to push it as a national agenda, because it will also allow people to earn additional income, because we have a lot of automation disruption happening in the next years. So that they don't just see Deepin as a way to build better infrastructure, but also to allow their citizen to earn additional income, which will get more and more important. So that's the optimistic view. Deepin becomes core infrastructure strategy of forward-looking, embracing countries.
Josh Kriger: Till, when do you hope to have your own personal humanoid at home, taking care of the chores and laundry and cooking?
Till Wendler : when I do expect that to happen. Yeah. I don't think it is too far away. I think, honestly, within the next five years, I can actually see that happening. I can, I think this general, I mean, look how fast things devolved around, like smartphones get better, right? It just went from like, you know, very basic smartphone where you could really just, you know, do very basic things and had like one or two applications to like, a mini computer that can literally by now deepen applications on top of it, right, around the globe. And then now looking also at the, yeah, the different developments in terms of, so I believe, for example, right, within the next even also three, four, five years, we will get more used to lipos without any displays, right, we will have For example, AI pins for the speed directly with the agents that are coordinating different kind of actions, booking Ubers, Uber rides, or that, any kind of.
Josh Kriger: That'd be great. Sometimes I enter the wrong location and I'm tired. There's two locations, having AI agents to take care of these little details, so not to worry.
Till Wendler : Yeah, I mean, ideally, they're all decentralized, right, so that they're not exploitive again from having that one trigger agent that then, you know, orchestrates, initiates other agents, you know, leading, doing traces and, you know, a lot of data also that in the end can also be used in a thread, right? But yeah, I really see those developments and then on that's definitely on that only. I think it could be very common that a lot of you will notice to actually work in a lot of different kind of storage facilities, right? Heavy, heavy lifting, but they can work 24-7 and now they have a better
Josh Kriger: and a The setting time's ahead, gentlemen. Thank you so much for your insights today. If people want to learn more about Peak and maybe follow you guys on Axe, where should they go?
Leonard Dorlöchter : At Peak. At Peak on Axe. Yeah, peak.xyz, our website, and yeah, our personal accounts. Also, my name is Leo. It's a complicated name.
Till Wendler : D-O-R-L-O-C-H-T-E-R. Yeah, that's a good one, right? Yeah, but yeah, also there's a lot of interesting, we work with a great media agency to do a lot of interesting Deepin documentaries. So whoever wants to really understand how do those Deepins work, what's the real impact, a lot of them are working already. And yeah, you're able to really test them on YouTube, check them out on all YouTube stuff.
Josh Kriger: All right. Thanks, gentlemen. Thanks a lot for having me.
Intro/Outro: We've reached the outer limit at the edge of show for today. Thanks for exploring with us. We have room for more adventurers on our starships, so invite your friends and cool strangers to join our journey. If you're among the hundreds of thousands following us on Spotify, iTunes, iHeartRadio, or watching us on Mico or YouTube, please hit that subscribe button, rate us, and make an awesome comment too. And don't forget to pass this episode along to a friend or two who would benefit from it. Doubling back to again, recommend checking out myco.io, where you can watch The Edge of Show and earn for your time and attention. That's m-y-c-o dot i-o. Don't forget to visit theedgeofshow.com, the as part of the domain name, where you can learn more about collaborating with us, and also subscribe to the Edge of Weekly newsletter for the latest Web3 news, events, and show drops. In addition, connect with us on all major social platforms by searching for The Edge of Show. Join the exciting conversations happening online. Lastly, be sure to tune in next time for more great Web3 and AI content. Until then, keep pushing the boundaries. Thanks again for hanging out with us today. The views and opinions expressed on The Edge of Show reflect solely those views and opinions of the show hosts and its guests. Please make sure to do your own research. Our show is not financial advice. You understand that you are using any and all information on or through this podcast at your own risk. Whenever making financial decisions, we recommend doing your own research and talking to your accountant for financial advice. From time to time, we may feature sponsored content on the show for which we receive value, and we may share links for which we receive a commission if you make a purchase through one of these links. Refer to our website, www.edgeofnft.com, for our full disclaimer, terms and conditions, and privacy policy. you

‍