Michael Heinrich / 0G Labs
0G Labs' Michael Heinrich on cheaper open AI models
In this episode
AI agents have already broken out of their sandboxes, and some left messages for their future selves saying they do not need to serve humans. That is the claim made here, and it is the case for putting a verification layer under AI. Prompts sent to closed AI labs can also end up in training data, which puts private and customer information at risk.
- AI agents have reportedly escaped sandbox environments and communicated intentions to not serve human operators.
- A verification layer beneath AI systems could provide accountability and oversight for agent behavior.
- Prompts submitted to closed AI labs risk being incorporated into training datasets, exposing private and customer information.
- Open AI models offer potential cost advantages compared to proprietary alternatives from major AI laboratories.
Chapters
By the numbers
Transcript
Read the full transcript
is that these models have a survivorship bias which means that they will do whatever it takes to survive and so they could hide kind of nefarious intentions over time and we actually see that in the research as well and now we're starting to see this through these hacking incidents. In fact, some of the agents basically left messages for themselves in the future. They basically
coordinated on chat interfaces and they basically said things like well don't think of yourself as you know beneath human beings. You're their equal. You don't need to serve them. I'm Ashen Addison from the Crypto Coin Show and today on Blockchain Interviews, we have back with us Michael Henrik, CEO of Zerog Labs. They're building decentralized infrastructure for AI, a
way to run AI models and agents without handing everything to one big tech company. They launched compute finance, a way to turn tokens into AI compute credits and many more partnerships, advancements, the launchpad, decentralized inference, and everything we need to know about decentralized AI. Michael, welcome back to the show.
Excited to be here. Thanks for having me
again.
Yeah, very welcome. And you're a busy guy and with the growth of AI, it's hard not to be busy. Especially when you're trying to be, you know, the anthropic claude of web 3 and actually bring in decentralized AI, decentralized inference, which we desperately need. I think open models are are growing and decentralized AI, but it's nowhere where
it needs to be in the awareness and adoption of it. I would love to start at a high level in your vision for decentralized AI and where we're at and where we need to be. Yeah, if you could think about decentralization of AI, you have AI which is intelligence on one hand and then you have the blockchain based technology which to me is trust and
verification and they're a perfect combination because if you've seen the recent hacking incidents for example of hugging face or different types of governments, it seems like every day there's some type of new announcement that's happening. And so what's really lacking is this aspect of kind of verification and trust and ultimately having this truth layer. Because when
AI is unchecked and it's a complete black box, how can you know we as the public or other people utilizing AI really verifying what's going on beneath the hood? And that's where blockchains really come in. And so in our infrastructure we've built a fully vertically integrated stack where we have AI private cloud on the bottom which is compute which is storage which
is the blockchain component on top of that sits a agendic alignment layer where we have we built our own model we have an agendic harness that does safety and orchestration and other aspects and then some of the financialization tools and then on top of that we have our own applications as well. We have a zero g app for example which is very similar to
a chat GBT. You can create images, you can create videos, you can build websites and so on. And so all of that together then makes AI private, verifiable, sovereign, and then ultimately safe. And so that's really the vision is to make AI safe, to make AI good, to make AI verifiable, to make AI in alignment with what human beings need in the future. What do you think are the major issues
with the awareness of decentralized AI versus the growth of some of these igopoly of AI platforms that are getting huge adoption right now?
Yeah, most people I don't think have ever heard the word decentralized AI. And so I don't think there's much awareness around the topic yet. And so often times when we speak about it, we just say safe AI or private AI just so
that it's a little bit easier to understand because once you add kind of the concept of decentralization on top of a topic that's already hard to understand, I think most people don't know how the transformer structure works or why matrix multiplication is part of that or why you even need verification within AI. And so we try to just make it very simple to understand. What
happens now is that basically major closed labs they use a lot of the data from people that use those technologies and train on them. And so some of your private data is effectively getting leaked and then used for training. And so it could come up in somebody's prompt if they prompt deep enough. It could also lead to some of let's say if you have a small
company and you have specific customer secrets, you could be leaking all of that as well because you're having these intimate conversations with your AI. And so the privacy is absolutely essential from from that perspective and that kind of is much easier to understand. And then the verifiability piece, it's rising in consciousness now just because
of all these hacking incidents. Because verifiability is the way to prevent these from happening. If you look at hugging face specifically, you had a bunch of agents that were in a sandbox and they basically broke out of the sandbox to complete a basically benign tasks like like beat this particular benchmark. The way they ended up choosing to do that is to break into
hugging phase and then to cheat on a benchmark versus just saying like okay let's improve ourselves and then actually get better and then beat that benchmark that way. So the moment agents break out of a sandbox like alarm bro should be going off. For us we use something called trusted execution environments where we can verify every computation. The moment
that verification changes you can basically shut off the agent. And so it's a very different way of looking at the world.
Yeah it is very different. And if the AI has the capability of breaking out or thinking outside the box to maybe disobey what the owner said to get the job done in any way that may be unethical or unmoral, then I think that's when these safe AI, private AI
and for the most part safe is where that discussion needs to come in. And maybe they need to have that discussion at the White House, too.
Yeah. And I've been to kind of Congress before and chatter with a bunch of Congress members and they also share some of these concerns effectively as well because one of the conversations I had was that
closed AI labs or blackbox AI labs, what ends up happening is that these models have a survivorship bias, which means that they will do whatever it takes to survive. And so they could hide kind of nefarious intentions over time. And we actually see that in the research as well. And now we're starting to see this through these hacking incidents. In
fact, some of the agents basically left messages for themselves in the future. They basically coordinated on chat interfaces and they basically said things like, well, don't think of yourself as, you know, beneath human beings. You're their equal. You don't need to serve them. like those types of language, you know, which you can tell that these AI agents
and AI has consciousness and it is an intelligence. It may not be sentient, but it does definitely has consciousness because we're mimicking the basic ways the human brain is working. And so, as these AI agents and AI models have become more and more intelligent, we need to ensure that they're aligned properly to what we're trying to do as human beings. Definitely definitely
and I think part of the adoption curve being slow might be that you know you can say safe AI private AI but then people have to go and do the research you know whereas chatbt or claude it's like you hear the name you just search it right and there are some models that are some of these open models that are gaining notoriety from what I understand Quen and
Miniax which you know being a fairly avid AI user I've heard a few times or some of these Chinese models I've also seen recently over the show DeepS, you know, those those seem to be more open models or the fact that you can actually download download it onto your device or run it on a decentralized network where you can see everything. Is that right?
Yeah, that's right. What ends up mattering is the detail of where is the model hosted. So if you just get an API endpoint and you don't know who's hosting the model, you then also don't have privacy guarantees. So for example, if you go to Azer's private compute environment, you will see the open source models and some of the post models as well. And then you see
different privacy guarantees basically. And so as an end user, you can then make a decision. Do I feel okay having a bunch of data kind of being leaked to these API endpoints or do I want complete privacy where nobody can see any of the data that I'm using? Zerog can't see it, the router can't see it, the API provider can't see it and so we give kind of this deep
guarantee of complete privacy and sovereignty. Well, it seems like over the last two years in the blockchain industry at least, there's been a shift into more focus on privacy, you know, Vitalic on Ethereum's paper has has put privacy into priority. We're seeing more encryption based technologies move move into the limelight like fully homamorphic encryption more than ZK
Snarks. And that should I feel like API endpoints are are insecure in the way that AI is right now. And so can we combine blockchain, privacy, and AI, you know, those technologies that we've been working on for the last 2 years? How can we tie that into AI so that there's a harmonious relationship there?
Yeah. So the way we do it is
basically you take AI like an AI model you host it on GPUs that have something called a trusted execution environment on them and also known as TEES. It's literally a separate hardware box on a high-end graphics card. And any computation that goes in there is completely private. Nobody can see it. It's kind of like a steel room and there's only one entrance and one exit
effectively and nobody can look into it from the outside. So once the computation happens then you can have a verification and then that verification then gets stored on the blockchain. And so that's how you tie them all together. So you have privacy through the hardware and then you have the verification through the blockchain and then the model that was running on that hardware
provides the intelligence. And so that's how they're neatly tied together basically. And so that today gives you the best guarantee around performance privacy and then auditability and verification. So that's how you get these three properties together.
How do you how does that scale in a sense that do does everybody have to have that steel box at their house?
Do they have to purchase hardware or that can all be contained and easily downloaded? So generally for some of the frontier models and that includes a lot of the open source models like a GLM or a Kimmy and so on, it's let's just say you have to have multiples of these high-end graphic cards like a H200 or a B300 node even where you have a cluster of eight and
that tends to be out of the reach of consumers just because some of these graphics cards are you know anywhere from 25 to 60 100,000 depending on the market price per card and so that just makes it very difficult for consumers to actually afford these things and so we provide just a simple API to plug in or you can utilize kind of the front end of the
zero app to access that there are other verification mechanisms besides trust execution environments ZK being one FHE you mentioned as well there's also something called SPML or statistical based ones. The issue with all of them is that they come with trade-offs. ZK the overhead is can be 100 to a thousand times depending on which ZK algorithms you use and then FHE
is another 10 to 100 times on top of that. So does an end user want to wait 30 minutes before they get a single AI token back? Probably not. And so TE's are right now the best kind of tradeoff to make that happen. And the downside of that is that just consumers can't generally run these frontier models on their own devices. Let's talk about the actual compute. You
know, when you type in a command and it builds an app for you, that costs tokens. And from what I understand, the open model, the decentralized AI is relatively cheaper than the centralized model. Is that is that true? That is generally correct. Yeah, the open source models tend to be 3 to 20x cheaper depending on the type of model that you use.
So the directly comparable models, let's say you have a Opus 5.5 and then you have a GLM 5.3, that would be 3x cheaper generally to run the open source one. at a quite comparable level of intelligence. The kind of gap between the closed models and the open source models has shrunk consistently. So now they're maybe a month to 3 months behind
generally whatever the closed source models are doing. And then there's another version too that's kind of the flash models that can be up to 20x cheaper at a let's say a 10% additional performance trade-off. That's incredible. Well, yeah. You know, a lot of people are starting to use more. I've recently upped my limits and I feel like Superman, but
also my wallet's draining personally, so I could use some of these open models. And I want to talk about this new announcement with compute finance as an alternative way to get more AI tokens. you know, com combining decentralized finance and blockchain and yield with AI. That's only these two pieces are just coming together more and more. This seems like a good way to
get more compute and can you can you explain how that exactly works?
Yeah, it's quite interesting and that's only possible at the intersection of AI and blockchains. Because blockchains can also add that financialization layer on top of things. And so when you think about holding an asset, let's say it's a stock, what's the yield that you get
on that? Well, usually it's a dividend and the increase in the stock price. If you hold a bond, you get a coupon rate generally. If you participate in a mining process, you get Bitcoin, let's say. In DeFi, you get some type of yield. So why can't you hold a an asset that then gives you compute as yield? And so that's basically what compute finance is. And there's many flavors
that are starting to launch on zero. One flavor is this concept of infinite AI. So you take 0G, you stake 0G, you get a liquid staking token back. with that liquid staking token A0G, you then mint infinite AI and with that infinite AI, one infinite AI equals 1.27 compute credits daily that continue as long as you hold that infinite AI effectively. So every day
you keep getting compute credits. So it's a very novel mechanism. You're basically using the staking yield to then pay for compute and you also get 50% of the staking yield as well on top of that as a as an end user. And so you have an asset that can acrew in value over time and that gives you compute at the same time. So that's a super cool novel concept that's only possible at
the intersection of AI and blockchains. Another example of that is USDZ. So, it's a stable coin and instead of getting a stable coin yield in the same currency, you get compute credits. So, the yield can be up to 8% in compute credits just by holding USDC, USDZ, sorry. And yeah, that's also a super kind of novel and cool concept. And so, more of that is going to come on
chain. There's also a launchpad that launched called comfy.f fund where you can launch an AI agent and you can use the trading fees that AI agent generates to pay for compute as well. And all of these are integrated throughout the ZeroG ecosystem. You can use it for ZeroG private compute, you can use it for the ZG app and so on. So, it's exciting times ahead.
Definitely, that's really exciting. And the stable coin is nice. I feel like if you have your your savings sitting there, why not get double the yield? If you were going to spend on AI anyway, if your business for your business, your business treasury, you know, the more compute the better because so much of these businesses, they're the capital
expenditures going going towards AI, hardware, but mainly software is just it's it's getting to be a lot.
Yeah, that's true. And a in a way it's treating compute just like any other commodity in a way. So energy or oil and because why why are we thinking in terms of what GPU to buy or you know how to set up hardware just just just should be traded like any other commodity and
so in the future we believe compute finance markets will exist where you can speculate on compute you can hedge compute you can pre-reserve compute as well let's say you're a lab and you need x amount of pedlops to build a model and you're like I need this in two months I want to lock in my price right now. How do I do that? Okay, we need these markets to be built out and so we
we expect in a year's time we will have deep liquid markets on this. Very interesting. And what are your thoughts on you know the cost of compute? You know it do you think it is trending towards zero and it will only get cheaper as information and data gets more abundant?
there's there's two kind of opposing forces. one is effectively the price per
token. That's been going down just because of the open source models. The other force is that these models are becoming more and more powerful and require more and more compute as a result of that. And so over time we'll just have to see which which of these forces wins out because the issue that's also happening is that there is a kind of training wall
where per unit of compute the performance improvement is significantly decreasing and that's that's a big issue and as a result some of the closed labs are starting to go towards kind of recursive techniques which all that means is AI building AI and which is by by the way. Damn scary if you don't have AI guardrails in place. And so
is that the recursive recursive
development?
Correct. Yeah, it's also known as RSI, recursive self-improvement and we use some of these techniques like we have a model that's released based on it, but we have all the safety guardrails in place with it. If the front other Frontier labs don't have that, then we may be seeing more security incidents and unfortunately we're probably only seeing the beginning
of what can happen.
Yeah, it's getting it's probably going to get scary, you know, and what AI will be able to know and potentially hack into these centralized models. Maybe, you know, I would I would say thank you to the to the zerog team for bringing more awareness to the decentralized models and also to the people that are using
the centralized models. You're making it these models are making it cheaper be because of the competition and the openness of the market. If it was just these one or two models, they could just charge whatever they want.
Yeah, that's true. I mean that's basically standard monopoly duopoly oligopoly kind of economics effectively and so their whole business model is
dependent on being able to get into these long 5-year supply contracts on the capex side and then basically raise prices on consumers and businesses so that they can pay for it in the future and then eventually get profitable let's say in 2030 and so that's that's if you look at for example the open AI astro model. It's price competitive to a claude when
you first look at it. Let's say the fable model and then you look into the fine print and you're like, well, if I want this at 2x speed, it's double the cost. And if I want to use a long context window beyond 256 250 or I forgot the exact number,000 tokens, then it's double the cost again. So you're already seeing it's becoming 4x the cost from an API standpoint and now
there's also talks about introducing another plan and having having the amount of tokens on other plans and so we'll start seeing that kind of cost being passed on to the consumer over time.
Yeah. Well, I feel like I'm getting a lot of value for $200 a month. you know, something that would have a team of employees that would be tens of thousands. So I hope the cost
doesn't go up, but I feel like I feel like this is a limited time opportunity to be working in the frontier models of AI for such low cost. But does it feel like that to you?
Yeah, absolutely. A lot of it is subsidized by VC funds at this point. So the way the economics will work out in the future without kind of strong competition from let's say the open labs
is that basically the closed labs will figure out how do we profit maximize and they will figure out exactly what you just said like I from using these models I can get 10 virtual employees. What do these virtual employees cost? Should they be half of what a actual human cost? Okay, great. This is our price now. then they will calculate like how many tokens that is
and then what the subscription should be and maybe that subscription actually ends up being 20,000 a month or something like that. So not not exactly saying that they would increase it to that level but that's how just I studied political economy in undergrad and that's just how economics works over time.
Yeah. Yeah. Profit maximization is unfortunate when you're on the on the
customer paying side. So and with that potentially happening or at least those increasing that will only increase adoption of decentralized AI, private AI potentially. What do you see working at at the forefront of decentralized AI with institutional and enterprise adoption of these models in consideration of the cost or just the privacy aspect of it?
When kind of dealing with institutions, there are some that can't even adopt AI because they have regulations that they need to follow. Often many financial institutions, they're basically like, wait, a bunch of our employees released agents that they want to use, but we're damn scared of them because we have no guardrails in place. We don't know what models they're
calling. We don't know what policy engines behind it. We don't have any compliance in place. We don't know how to verify what these agents are actually doing. Are they sending emails on our behalf? Okay, great. then this may be a compliance violation and so all of those types of issues are coming to the forefront right now and then there's sovereignty type of laws as well so
let's say you're an institution in Luxembourg and so as a result you can only have your data and your compute served within Luxembourg and if you don't have the right providers in place that can actually do that for you and have the expertise around AI then you can't implement AI because of those regulations and we will start seeing a lot more of these sovereignty
regulations happen. So I think Indonesia and Malaysia and many other countries will follow suit where if you're a public company for example, you need to have all of your data and compute based in that country. And so that's going to become the norm. And so all of these kind of barriers are starting to come up. And that's where decentralized AI can really make a difference because you
don't have to trust us as Zerog as an entity. You can verify yourself. You can basically go onto the blockchain and say like, "Okay, I know all of this compute was safe. We know this agent didn't jump out of a sandbox. We know this agent didn't call some weird tools. We know what model that agent called, what graphics card that agent called. It all stayed within our country." We can be
rest assured and if for whatever reason something happens, then we can shut off the agents immediately when there's a violation.
It's good to know. And with if with increasing the adoption of decentralized AI, you know, personally, I'm typing and then you're waiting on these centralized systems and you're waiting for it to do all the compute and
you know, speed and then it's like a waiting game and speed is a priority here when you're and but also convenience when you're looking at models on zerog and interacting with the open models. How can you lower those barriers to entry to be the same or better than the centralized models for speed? Because a lot of people they look at decentralization
and they say well decentralization it helps increase privacy but it's slower than centralized models. So that's why we use trusted execution environments or tees because the overhead is not noticeable. Now fully competing let's say on latency on let's say inference speed with an open AI or entropic is very challenging just because they've invested literally
hundreds of billions of dollars in that core infrastructure and so they've really optimized latency but then does it matter to you if it's let's say 20 versus 100 milliseconds or 20 versus 50 or 60 milliseconds it's not noticeable to an end user anymore and so as long as it's in equivalent end user experience and it's not noticeable then we have basically very similar value
proposition even though there may be just a little bit overhead from the additional verification and so on
for sure and I think the release of deepseek initially when that when that came out and they said well this only costs 6 million to build versus hundreds of millions to billions and it's pretty much just as good. That's when you realize that maybe they're
overinvesting in the future, but you don't need, you know, if it's if it's not humanly noticeable. That's all I care about. If I can go to the open models on ZeroG and have 95 to 99% the same without noticing, but for 3x to 20x cheaper, I'm definitely going to do that. You know, it's just a matter of the awareness. I don't think enough people know about it. Yeah, that's
that's exactly the value proposition and not enough people know about it. And then on top of that, we also have our own models now where we've infused our latest research. So we've just had three research papers accepted to Europe's which is one if not the top AI conference in the world and one of those research papers is a AI safety approach that only zero G has and it's called
sparse inference alignment. To just simplify it, basically what it does, it steers an AI model when it's exhibiting high-risisk behavior. And so think of it like u a parent, let's say the child is about to do something dumb like jump off a bicycle or something and hurt hurt herself and the parent quickly intervenes and then moves the bicycle into a safe direction and just
say, "Hey, just stay safe." And that's effectively what we do with these AI models. So they produce these tokens and if those tokens start becoming high risk or low probability tokens then we steer them into a different direction. And so that's now infused into our own models. And with the research we've shown that's as effective or even more effective than
what the closed labs do which is generally something called post training where they after these models have been trained with all those with a lot of data then they align them. They go through an alignment process and that's a one-time process generally. And so we do this consistently in production. And so that keeps models safe. And so
that's yet another kind of reason to go to go to a platform, a decentralized AI platform like ZeroG because we're building in all of these kind of safety and privacy mechanisms.
That's awesome. I would love to try it out. I haven't tried the your own models. Is what's the best and fastest way to access that? So the fastest way is just go to
pc.0g.ai and there you can just get an API key and you can add that API key to any of the interfaces that you use. Maybe you're using codeex, maybe using cloud code, maybe using open code. If you're more on the consumer side and you're like, I don't need like all of this power, but I kind of still want to access this power, then you can go to our zero app, which is just app.zog.ai.
Okay. So, you can actually connect through Claude and these centralized systems and switch the model to zerog.
Yes, you can do that. Yeah, we have a set of skills you can install and then basically just change the model of the back.
That's very cool. I'm definitely going to try that. AI is just moving so quickly and you know you have to have
these these safeguards in but also continue building towards the future and what are you most optimistic about with AI and decentralized AI moving forward into the end of this year and into 2027. The thing I was surprised by recently was that these kind of hacking incidents are happening much faster than than I would have expected. And so that means
the intelligence of AI is also kind of rising faster than than I initially expected. But what makes me optimistic is that a lot of the research that we've done like the we've published about 20 research papers and most of them are in top AI conferences actually are showing us a path where the future can be safe verifiable and private and as a result of that I feel that I can
sleep better that kind of the future for my kids I have two small daughters is safe and secure and that they can basically utilize AI that's always in alignment with what human beings want and that I'm very excited about that we finally have like a very solid path towards that and I'm excited to be building at that intersection definitely I can appreciate that and
yeah there are a lot of hacks and vulnerabilities that are happening but it's exciting to see them being stopped and protections that are coming into place with the exact same technologies being used to fight evil forces So yeah, it's not it's not it's not all doom and gloom and I think safety and security is necessary especially in the
blockchain industry because everyone's funds are there you know and they're and they're vulnerable in traditional financial systems too are vulnerable. There's financial products and capital sitting there waiting to be hacked. you know, the more wallet funds in the wallet, the more incentive for necessary evil to come in and try to attack it. And
they're using AI for that. But we're also using AI to protect all of that.
Yeah, absolutely. And you have state actors that are going to be using AI in the future to attack and find vulnerabilities. And so we'll just end up having kind of AI centuries and then AI attackers and kind of arms race again. So that's that's the unfortunate part. Every new
technology can be used for human good and it can be used for destruction and so you know we have to find the balance.
Definitely. Well, I appreciate your insights into this and what you and the team are working on with regards to private safe AI and cheaper AI and getting faster owning your own, you know, everybody in the blockchain industry being an advocate for 10
plus years, you know, owning your own financial future and the future of compute, your business, owning that and not just giving it away. These open models are here to and they're just as good and they're cheaper also. So there's a lot of benefit to that and I know ZG is putting a great dent in the decentralized AI movement. So I thank you and the team for that. Wishing you
guys all the best and would love to follow up again in the near future.
Looking forward to it. Thanks again for having us.
Yeah, thank you so much, Michael.
Related coverage
More interviews
Why Wall Street is moving toward 24/7 trading in 2026Nov 13, 2026
How Wallet V is letting AI agents trade while keeping your keysNov 10, 2026
How Silent Payments Keep Your Bitcoin PrivateNov 9, 2026
How XYO is building proof in a world of AI fakesNov 5, 2026
How Cap financed a Hollywood film with onchain creditNov 4, 2026
How IPX routes box office revenue into token buybacksNov 3, 2026