My short answer to whether DeFi belongs on ICP is yes. I think ICP is the safest DeFi environment in crypto, and I think most of the rest of DeFi is more fragile than people want to admit. The four posts on this page make that case from different angles: how dangerous the current DeFi stack is, what fully on-chain exchanges mean for centralized ones, and what I think Multidex shows about what ICP can do. None of it is financial advice, and anything can happen, but this is where I stand.
The risk is in the plumbing. A typical Ethereum DeFi action can pass through a browser wallet, a website and several other third parties before it reaches the chain, and each one is a place to get hacked, phished or drained. On ICP, I argue, you can hold assets from other chains in a wallet like OISY and use protocols such as Liquidium without bridges or a company in the middle. Liquidium's loans are over-collateralized and the protocol is non-custodial. There are still risks, and you can still lose a loan if the price drops, but I would estimate that building on one chain cuts the risk by 90-plus percent. That is my estimate, not a measurement.
My view started with a warning. In When Ethereum DeFi Collapses, Money Will Flow Into ICP, from June 14, 2026, I argued that Ethereum DeFi is a recursive house of cards, with people looping liquid-staking tokens from one protocol into the next, so one AI-assisted hack could cascade through all of it. I labeled that as a prediction. The practical side was the alternative I pointed to: locking ICP on the Network Nervous System, which I described at about 2.22 percent for a two-week lock and about 6.65 percent for a two-year lock, with votes that actually steer the network.
The next day, in Crypto Exchanges Are Threatened by ICP, I moved from DeFi to the exchanges themselves. My argument was that exchanges are the real power in crypto, and that a chain where you hold your own assets and interact with it directly removes most of the reason to use them. I also repeated the point that stuck with me from an analysis by Bobby O: ICP listed on major exchanges at hundreds of dollars, roughly 100 times the last funding round, and then fell about 99 percent. I present that as one reading of what happened, not settled fact.
In July I put numbers on the comparison. In Crypto DeFi Is Broken. ICP Fixed It, I argued that the rest of DeFi is dangerous enough that people keep losing money to compromised browsers and bad links, and that ICP's model avoids most of that. I also built a scorecard of 100 yes-or-no questions about what a blockchain can do. I scored ICP at 97 yes answers and Ethereum at 22. That is my own scorecard, so judge it on its questions, but it is the reason I think ICP can do almost everything Ethereum can, plus a lot more.
By August the argument had a product. In ICP Can Run an Entire Crypto Exchange On-Chain, from August 21, 2026, I pointed to Multidex at multidex.ai, which I described as an exchange whose matching engine, order books, custody, prices and front end are all canister code on a public network. I contrasted that with Hyperliquid, which I called closed source, and with dYdX, where matching happens off-chain. I also described a treasury that pays for its own compute and burns ICP once it passes a set threshold. That is my reading of the project's documentation and stated design, not an audit.
What you will find below is all four posts in full, listed newest first in the contents, so you can read the most complete version of each argument. If you want the practical side of holding ICP while you wait, my NNS staking tutorial covers how locking works, and the bear case for ICP covers the strongest arguments against my view. I would read that second one before you act on anything here.
If you are holding ICP and want my read on your own position, you can book a personal video review.
On this page
Every post I have written on this subject is below in full, newest first, each under its own heading.
- ICP Can Run an Entire Crypto Exchange On-Chain (August 21, 2026)
- ICP DeFi and On-Chain Exchanges: Why I Think ICP Wins (July 9, 2026)
- Crypto Exchanges Are Threatened by ICP (June 15, 2026)
- When Ethereum DeFi Collapses, Money Will Flow Into ICP (June 14, 2026)
ICP Can Run an Entire Crypto Exchange On-Chain
From my video published August 21, 2026.
Internet Computer Protocol can run an entire crypto exchange 100% on-chain. That is an absolute game changer, and Multidex is showing this off at multidex.ai. In my view, this is one of the single biggest reasons to be hyped about ICP right now. Let's dive into this and see what it means, because if you've seen that DFINITY is putting Multidex out, you might not understand what a big deal it is.
Multidex Versus the Rest of Crypto
First, let's compare what Multidex can do versus what the rest of crypto is doing. Obviously, centralized exchanges are just garbage. You are having to trust them for everything, there's no transparency, and there are enough centralized exchanges already that we don't need any more. Right now there's been a lot of excitement about Hyperliquid, dYdX, and Ethereum DeFi. Multidex is so vastly superior to all of that, in my opinion, that there's huge value that can come from it. So let's look at the difference between "100% on-chain" and what a true decentralized exchange actually means.
"On-chain" is a claim. This is straight from the Multidex documentation at multidex.ai, under the docs and launch section, where you scroll down and read it: on-chain is a claim almost everyone makes and almost no one fully meets. Multidex means it literally. The matching engine, the order books, the custody, the prices, and the front end you're reading are all canister code on a public network. That is so insanely bullish for ICP. What's amazing is that DFINITY just built this with AI as one of the many things they're doing. So let's compare what they quickly built with AI, and could be launching any time now, against Ethereum DeFi.
Why Ethereum DeFi Front Ends Are a Liability
On Ethereum DeFi, the contracts may be on-chain, but the front ends aren't. The websites run on for-profit companies' ordinary servers, which can change, gate users, or disappear. Why does that matter? If you've tried to use things like Hyperliquid or various Ethereum DeFi exchanges, you may have noticed that if you're in certain countries, you get restricted. A truly decentralized exchange that is fully on-chain is not going to country-restrict you, and it's not going to suddenly change or disappear either.
Because the website with Ethereum DeFi is where you actually sign transactions, if anybody hacks the website, the traders are directly at risk. With AI hacking, websites being hacked, especially anything involving Ethereum DeFi, are extremely lucrative targets to hack and make a lot of money from. That makes those sites expensive to maintain and secure, with a high probability of a loss, whether that's the loss from how expensive and difficult it is to maintain and secure against AI hacking, or the loss when it actually gets hacked. You have all these different vectors: a compromised domain, an injected script that can cause you to lose money even when you're on the official website, not to mention all the phishing attacks using fake versions of the same website.
Here, with the app on ICP, Multidex, which is in play mode right now, can have the exchange served directly by the canister. There's no company-run website in the middle to compromise. On Ethereum DeFi you can also get front-run by validator nodes ordering transactions, which is what's happening a lot under the name miner extractable value. On Multidex the exchange is served directly by the canister instead.
Hyperliquid: Closed Source Is Not Decentralized
So let's compare now to Hyperliquid, which has gotten an undeserved amount of hype, except for the branding. Look at the reality of it versus the gigantic market cap and all the hype. Somebody in my chat at jerrybanfield.com said that Trump just mentioned it, and who, how, and why that happened is sketchy and easy to predict. Who close to him has a bunch of money in it?
Hyperliquid is fast, but it runs on what amounts to a dedicated, purpose-built cluster and is entirely closed source. No one outside the company can see it. The traders must place their full faith in the operator. How, in anyone's mind, is that decentralized? When the code is closed source and you have to have full faith in them, it's ridiculous. You can't verify the matching rules or the fees, or whether the house quietly advantages itself. It does not run on a general-purpose network, whereas Multidex runs on a general-purpose network and is open source. So Multidex on ICP is a real version of what Hyperliquid is essentially pretending to be: decentralized, but not actually offering it when you dig into the details. So typical for crypto.
dYdX: Off-Chain Matching With No Transparency
Then compared to dYdX, the order matching happens off-chain in an engine the operator runs, and only the settlement touches the chain. So the rules that decide whose order gets filled first and in what order are not public, and there's no way to have transparency around that. Multidex matches fully on-chain with rules that use public code and orders that are staged, so there's no queue to jump and release priorities aren't invisible.
This is so technically superior to everything else that exists that I think it's a clear choice for anyone who truly wants the best trading experience. The question is, who cares about the best trading experience? Are people just going off marketing? Well, there's definitely a subset of traders who will find these properties extremely attractive. And the more problems there are with these other exchanges, whether it's getting hacked, people getting ripped off, transactions being front-run, or trust issues, all of these are full of potential problems and reasons they could dump and go to zero, and any of those problems will feed straight into Multidex.
What Multidex Means for ICP Holders
Now for ICP holders, here's a key point. The exchange collects fees into a treasury that first pays for its own compute, so it fuels itself when no operators subsidize it. Beyond that, the design intent is public good. Once the treasury exceeds a set threshold, further gains are used to burn the ICP token, reducing supply and benefiting every ICP holder. The exchange doesn't enrich an owner, whereas Hyperliquid is massively enriching its owners. The entire value of all the publicly held Hyperliquid tokens is a fraction of what the owners and insiders already have. Meanwhile, Multidex is burning ICP, and to govern it, you'll be voting on the Network Nervous System. So ICP becomes the governance token for Multidex on top of everything else ICP already is.
This was written by AI, and it shows the properties of the ICP infrastructure that are special, because someone else could come build the exact same thing if they wanted to. Instead of putting it straight on the NNS, they could put their own token behind it. So this is a proof of concept for somebody else who might have some money, want to market it, and profit off it. They may just build their own version of this directly on ICP too. That's just another reason today to be insanely bullish on ICP. This is so special, and so far past what anybody else is doing in crypto.
You can see how disingenuous almost everything and everyone is in crypto, promoting an exchange like Hyperliquid as decentralized when all the code is closed source, there's almost no transparency, and insiders are making almost all the money, while all these people voluntarily do marketing for it and talk about how great it is. It's insane. ICP allows anyone to just build an exchange like this too, and DFINITY has simply led by example.
From Technology to Products
What is special about ICP? Multidex is one single product. Dominic said a while ago that the first thing they had to do was get the technology to a critical threshold where it can do everything you need it to do. For most of a decade, DFINITY poured most of its resources into advancing the tech. Over the last year or two, there's been a shift, where now the tech is good enough to do almost anything you want it to do. Now you need products. Multidex is a product that could be huge. Caffeine AI is another product. Cloud engines, another product. This is spiraling, this is snowballing, where there are so many products and so many ways to win. Something amazing is going to happen.
I Love a Leaderboard
Here's an example of something that's awesome because of this setup. I love a leaderboard. Multidex right now has just a play leaderboard, but can you imagine what that becomes? I just talked to a guy yesterday who is number one on the leaderboard on my website now, and I'm very grateful. He came in and dropped about $300 in a single day on jerrybanfield.com. He paid for the chat at jerrybanfield.com/chat, and then he went in and paid for one of my talk-to-me-now options, where you can talk to me right now for 10 minutes. He bought that to start, and then he was having a good time, so he bought the talk-to-me-now for 30 minutes twice more after that. Now he's number one on the leaderboard because he spent $327 on jerrybanfield.com yesterday, which I'm very grateful for. He found me through my videos on X.
Where I'm going with that is: imagine someone like him, who trades a lot, having a platform where he can show off his trading profit against everybody else who's trading. Because it's on-chain, you can set things like this up and show off a leaderboard. And keep in mind, anyone could use AI to build something like this too.
So for example, if I were to, which I don't, because I'm a content creator and I'm in Florida with regulations and all that stuff, but imagine anyone else who maybe has some money lying around, a few million, and they look at Multidex and say, "I'm going to build my own version of this, it's going to have my token behind it, and I'm going to pay people to market it." Somebody could just build their own. You don't understand how difficult it is to build something like this normally. I will give Hyperliquid and dYdX credit on this point, because when you're using all that off-chain infrastructure and all these websites, there's so much security and so many things you have to address. You can use AI coding, but then you also have to do a whole bunch of coding and solve a whole bunch of problems with real developers. Because Multidex is all on-chain, you can just build all of this with AI. That means somebody else may come in and build this too.
So on ICP, it may become the chain. It may take off because everybody starts building their own version of Hyperliquid, or dYdX, or Multidex, all on ICP. And here's the amazing thing: ICP is set up so you can use native Bitcoin, Ethereum, and Solana. You don't have to stop using all those other tokens. You can just build all that stuff, stablecoins included. It's all sitting there, and you can build it with AI. So Multidex is something that by itself could be huge, and it's proof of what ICP can do, which is huge. People who want to make money are going to see this, look at it, and prepare to make themselves rich. To me, the future is built on ICP, and Multidex is a very good indication of what's possible.
All Roads Lead to ICP
Other people are going to copy it and build their own versions, because it's just too easy to use AI to make your own version of Multidex, and there are huge amounts of money somebody can make off it. The leaderboard is just too incredible. It's a feature that would make a lot of money and would attract people to actually trade on it, because they'll want to show off. Imagine if you had all this profit and, even if you didn't show anyone publicly, you could just sit there and check the leaderboard and compete with all these other people on the money. Like the guy I talked with yesterday, that's someone who might really appreciate being able to do that. So there's a clear reason why people would move over to Multidex.
ICP feels like the game is rigged, like we are the house in the casino. All these other coins and crypto are the casino, and all roads are leading to ICP. This is why I'm all in on ICP. It's why I don't hold anything else, and why I don't care what happens with anything else in crypto. This is where the action is going. This infrastructure is incredible, and Multidex is one example of that. This obviously wasn't financial advice, and obviously I'm an ICP maxi, and this is all I do.
If you'd like to chat with me and pop up on the leaderboard on jerrybanfield.com, you'll notice that my entire website is built with AI on the ICP blockchain too. You can join the chat and talk with me and everybody else in there.
You can check out my Officer Banfield book.
I do dating coaching as well.
I do YouTube coaching, and I have one-on-one calls. If you see that I'm online, like I was today, you can just jump into a call. The call yesterday started within 15 minutes or so of when the buyer paid, and he paid for everything in ICP.
If you've got a crypto video or project you're working on, or you've built something on ICP you want to show off, I'll do a sponsored crypto video with you.
I've also got my Skool community.
And there are tips if you want to show off on the tip leaderboard as well.
Thank you very much for reading. For more on why I'm all in, see my ICP Crypto playlist. Let's get excited, and let's stay excited every day about ICP.
ICP DeFi and On-Chain Exchanges: Why I Think ICP Wins
Originally published July 9, 2026.
My short answer to whether DeFi belongs on ICP is yes. I think ICP is the safest DeFi environment in crypto, and I think most of the rest of DeFi is more fragile than people want to admit. The four posts on this page make that case from different angles: how dangerous the current DeFi stack is, what fully on-chain exchanges mean for centralized ones, and what I think Multidex shows about what ICP can do. None of it is financial advice, and anything can happen, but this is where I stand.
The risk is in the plumbing. A typical Ethereum DeFi action can pass through a browser wallet, a website and several other third parties before it reaches the chain, and each one is a place to get hacked, phished or drained. On ICP, I argue, you can hold assets from other chains in a wallet like OISY and use protocols such as Liquidium without bridges or a company in the middle. Liquidium's loans are over-collateralized and the protocol is non-custodial. There are still risks, and you can still lose a loan if the price drops, but I would estimate that building on one chain cuts the risk by 90-plus percent. That is my estimate, not a measurement.
My view started with a warning. In When Ethereum DeFi Collapses, Money Will Flow Into ICP, from June 14, 2026, I argued that Ethereum DeFi is a recursive house of cards, with people looping liquid-staking tokens from one protocol into the next, so one AI-assisted hack could cascade through all of it. I labeled that as a prediction. The practical side was the alternative I pointed to: locking ICP on the Network Nervous System, which I described at about 2.22 percent for a two-week lock and about 6.65 percent for a two-year lock, with votes that actually steer the network.
The next day, in Crypto Exchanges Are Threatened by ICP, I moved from DeFi to the exchanges themselves. My argument was that exchanges are the real power in crypto, and that a chain where you hold your own assets and interact with it directly removes most of the reason to use them. I also repeated the point that stuck with me from an analysis by Bobby O: ICP listed on major exchanges at hundreds of dollars, roughly 100 times the last funding round, and then fell about 99 percent. I present that as one reading of what happened, not settled fact.
In July I put numbers on the comparison. In Crypto DeFi Is Broken. ICP Fixed It, I argued that the rest of DeFi is dangerous enough that people keep losing money to compromised browsers and bad links, and that ICP's model avoids most of that. I also built a scorecard of 100 yes-or-no questions about what a blockchain can do. I scored ICP at 97 yes answers and Ethereum at 22. That is my own scorecard, so judge it on its questions, but it is the reason I think ICP can do almost everything Ethereum can, plus a lot more.
By August the argument had a product. In ICP Can Run an Entire Crypto Exchange On-Chain, from August 21, 2026, I pointed to Multidex at multidex.ai, which I described as an exchange whose matching engine, order books, custody, prices and front end are all canister code on a public network. I contrasted that with Hyperliquid, which I called closed source, and with dYdX, where matching happens off-chain. I also described a treasury that pays for its own compute and burns ICP once it passes a set threshold. That is my reading of the project's documentation and stated design, not an audit.
What you will find below is all four posts in full, listed newest first in the contents, so you can read the most complete version of each argument. If you want the practical side of holding ICP while you wait, my NNS staking tutorial covers how locking works, and the bear case for ICP covers the strongest arguments against my view. I would read that second one before you act on anything here.
If you are holding ICP and want my read on your own position, you can book a personal video review.
On this page
Every post I have written on this subject is below in full, newest first, each under its own heading.
- ICP Can Run an Entire Crypto Exchange On-Chain (August 21, 2026)
ICP Can Run an Entire Crypto Exchange On-Chain
From my video published August 21, 2026.
Internet Computer Protocol can run an entire crypto exchange 100% on-chain. That is an absolute game changer, and Multidex is showing this off at multidex.ai. In my view, this is one of the single biggest reasons to be hyped about ICP right now. Let's dive into this and see what it means, because if you've seen that DFINITY is putting Multidex out, you might not understand what a big deal it is.
Multidex Versus the Rest of Crypto
First, let's compare what Multidex can do versus what the rest of crypto is doing. Obviously, centralized exchanges are just garbage. You are having to trust them for everything, there's no transparency, and there are enough centralized exchanges already that we don't need any more. Right now there's been a lot of excitement about Hyperliquid, dYdX, and Ethereum DeFi. Multidex is so vastly superior to all of that, in my opinion, that there's huge value that can come from it. So let's look at the difference between "100% on-chain" and what a true decentralized exchange actually means.
"On-chain" is a claim. This is straight from the Multidex documentation at multidex.ai, under the docs and launch section, where you scroll down and read it: on-chain is a claim almost everyone makes and almost no one fully meets. Multidex means it literally. The matching engine, the order books, the custody, the prices, and the front end you're reading are all canister code on a public network. That is so insanely bullish for ICP. What's amazing is that DFINITY just built this with AI as one of the many things they're doing. So let's compare what they quickly built with AI, and could be launching any time now, against Ethereum DeFi.
Why Ethereum DeFi Front Ends Are a Liability
On Ethereum DeFi, the contracts may be on-chain, but the front ends aren't. The websites run on for-profit companies' ordinary servers, which can change, gate users, or disappear. Why does that matter? If you've tried to use things like Hyperliquid or various Ethereum DeFi exchanges, you may have noticed that if you're in certain countries, you get restricted. A truly decentralized exchange that is fully on-chain is not going to country-restrict you, and it's not going to suddenly change or disappear either.
Because the website with Ethereum DeFi is where you actually sign transactions, if anybody hacks the website, the traders are directly at risk. With AI hacking, websites being hacked, especially anything involving Ethereum DeFi, are extremely lucrative targets to hack and make a lot of money from. That makes those sites expensive to maintain and secure, with a high probability of a loss, whether that's the loss from how expensive and difficult it is to maintain and secure against AI hacking, or the loss when it actually gets hacked. You have all these different vectors: a compromised domain, an injected script that can cause you to lose money even when you're on the official website, not to mention all the phishing attacks using fake versions of the same website.
Here, with the app on ICP, Multidex, which is in play mode right now, can have the exchange served directly by the canister. There's no company-run website in the middle to compromise. On Ethereum DeFi you can also get front-run by validator nodes ordering transactions, which is what's happening a lot under the name miner extractable value. On Multidex the exchange is served directly by the canister instead.
Hyperliquid: Closed Source Is Not Decentralized
So let's compare now to Hyperliquid, which has gotten an undeserved amount of hype, except for the branding. Look at the reality of it versus the gigantic market cap and all the hype. Somebody in my chat at jerrybanfield.com said that Trump just mentioned it, and who, how, and why that happened is sketchy and easy to predict. Who close to him has a bunch of money in it?
Hyperliquid is fast, but it runs on what amounts to a dedicated, purpose-built cluster and is entirely closed source. No one outside the company can see it. The traders must place their full faith in the operator. How, in anyone's mind, is that decentralized? When the code is closed source and you have to have full faith in them, it's ridiculous. You can't verify the matching rules or the fees, or whether the house quietly advantages itself. It does not run on a general-purpose network, whereas Multidex runs on a general-purpose network and is open source. So Multidex on ICP is a real version of what Hyperliquid is essentially pretending to be: decentralized, but not actually offering it when you dig into the details. So typical for crypto.
dYdX: Off-Chain Matching With No Transparency
Then compared to dYdX, the order matching happens off-chain in an engine the operator runs, and only the settlement touches the chain. So the rules that decide whose order gets filled first and in what order are not public, and there's no way to have transparency around that. Multidex matches fully on-chain with rules that use public code and orders that are staged, so there's no queue to jump and release priorities aren't invisible.
This is so technically superior to everything else that exists that I think it's a clear choice for anyone who truly wants the best trading experience. The question is, who cares about the best trading experience? Are people just going off marketing? Well, there's definitely a subset of traders who will find these properties extremely attractive. And the more problems there are with these other exchanges, whether it's getting hacked, people getting ripped off, transactions being front-run, or trust issues, all of these are full of potential problems and reasons they could dump and go to zero, and any of those problems will feed straight into Multidex.
What Multidex Means for ICP Holders
Now for ICP holders, here's a key point. The exchange collects fees into a treasury that first pays for its own compute, so it fuels itself when no operators subsidize it. Beyond that, the design intent is public good. Once the treasury exceeds a set threshold, further gains are used to burn the ICP token, reducing supply and benefiting every ICP holder. The exchange doesn't enrich an owner, whereas Hyperliquid is massively enriching its owners. The entire value of all the publicly held Hyperliquid tokens is a fraction of what the owners and insiders already have. Meanwhile, Multidex is burning ICP, and to govern it, you'll be voting on the Network Nervous System. So ICP becomes the governance token for Multidex on top of everything else ICP already is.
This was written by AI, and it shows the properties of the ICP infrastructure that are special, because someone else could come build the exact same thing if they wanted to. Instead of putting it straight on the NNS, they could put their own token behind it. So this is a proof of concept for somebody else who might have some money, want to market it, and profit off it. They may just build their own version of this directly on ICP too. That's just another reason today to be insanely bullish on ICP. This is so special, and so far past what anybody else is doing in crypto.
You can see how disingenuous almost everything and everyone is in crypto, promoting an exchange like Hyperliquid as decentralized when all the code is closed source, there's almost no transparency, and insiders are making almost all the money, while all these people voluntarily do marketing for it and talk about how great it is. It's insane. ICP allows anyone to just build an exchange like this too, and DFINITY has simply led by example.
From Technology to Products
What is special about ICP? Multidex is one single product. Dominic said a while ago that the first thing they had to do was get the technology to a critical threshold where it can do everything you need it to do. For most of a decade, DFINITY poured most of its resources into advancing the tech. Over the last year or two, there's been a shift, where now the tech is good enough to do almost anything you want it to do. Now you need products. Multidex is a product that could be huge. Caffeine AI is another product. Cloud engines, another product. This is spiraling, this is snowballing, where there are so many products and so many ways to win. Something amazing is going to happen.
I Love a Leaderboard
Here's an example of something that's awesome because of this setup. I love a leaderboard. Multidex right now has just a play leaderboard, but can you imagine what that becomes? I just talked to a guy yesterday who is number one on the leaderboard on my website now, and I'm very grateful. He came in and dropped about $300 in a single day on jerrybanfield.com. He paid for the chat at jerrybanfield.com/chat, and then he went in and paid for one of my talk-to-me-now options, where you can talk to me right now for 10 minutes. He bought that to start, and then he was having a good time, so he bought the talk-to-me-now for 30 minutes twice more after that. Now he's number one on the leaderboard because he spent $327 on jerrybanfield.com yesterday, which I'm very grateful for. He found me through my videos on X.
Where I'm going with that is: imagine someone like him, who trades a lot, having a platform where he can show off his trading profit against everybody else who's trading. Because it's on-chain, you can set things like this up and show off a leaderboard. And keep in mind, anyone could use AI to build something like this too.
So for example, if I were to, which I don't, because I'm a content creator and I'm in Florida with regulations and all that stuff, but imagine anyone else who maybe has some money lying around, a few million, and they look at Multidex and say, "I'm going to build my own version of this, it's going to have my token behind it, and I'm going to pay people to market it." Somebody could just build their own. You don't understand how difficult it is to build something like this normally. I will give Hyperliquid and dYdX credit on this point, because when you're using all that off-chain infrastructure and all these websites, there's so much security and so many things you have to address. You can use AI coding, but then you also have to do a whole bunch of coding and solve a whole bunch of problems with real developers. Because Multidex is all on-chain, you can just build all of this with AI. That means somebody else may come in and build this too.
So on ICP, it may become the chain. It may take off because everybody starts building their own version of Hyperliquid, or dYdX, or Multidex, all on ICP. And here's the amazing thing: ICP is set up so you can use native Bitcoin, Ethereum, and Solana. You don't have to stop using all those other tokens. You can just build all that stuff, stablecoins included. It's all sitting there, and you can build it with AI. So Multidex is something that by itself could be huge, and it's proof of what ICP can do, which is huge. People who want to make money are going to see this, look at it, and prepare to make themselves rich. To me, the future is built on ICP, and Multidex is a very good indication of what's possible.
All Roads Lead to ICP
Other people are going to copy it and build their own versions, because it's just too easy to use AI to make your own version of Multidex, and there are huge amounts of money somebody can make off it. The leaderboard is just too incredible. It's a feature that would make a lot of money and would attract people to actually trade on it, because they'll want to show off. Imagine if you had all this profit and, even if you didn't show anyone publicly, you could just sit there and check the leaderboard and compete with all these other people on the money. Like the guy I talked with yesterday, that's someone who might really appreciate being able to do that. So there's a clear reason why people would move over to Multidex.
ICP feels like the game is rigged, like we are the house in the casino. All these other coins and crypto are the casino, and all roads are leading to ICP. This is why I'm all in on ICP. It's why I don't hold anything else, and why I don't care what happens with anything else in crypto. This is where the action is going. This infrastructure is incredible, and Multidex is one example of that. This obviously wasn't financial advice, and obviously I'm an ICP maxi, and this is all I do.
If you'd like to chat with me and pop up on the leaderboard on jerrybanfield.com, you'll notice that my entire website is built with AI on the ICP blockchain too. You can join the chat and talk with me and everybody else in there.
You can check out my Officer Banfield book.
I do dating coaching as well.
I do YouTube coaching, and I have one-on-one calls. If you see that I'm online, like I was today, you can just jump into a call. The call yesterday started within 15 minutes or so of when the buyer paid, and he paid for everything in ICP.
If you've got a crypto video or project you're working on, or you've built something on ICP you want to show off, I'll do a sponsored crypto video with you.
I've also got my Skool community.
And there are tips if you want to show off on the tip leaderboard as well.
Thank you very much for reading. For more on why I'm all in, see my ICP Crypto playlist. Let's get excited, and let's stay excited every day about ICP.
Crypto Exchanges Are Threatened by ICP
From my video published June 15, 2026.
The Internet Computer is a massive threat to crypto exchanges, and most of us in ICP don't fully consider what a problem that creates. Once you see who actually holds the money and power in crypto, a lot of things start to make sense — including why ICP has been kept so quiet. This is my honest read on the space, not financial advice.
The exchanges are the real power
We tend to think: ICP has the best technology, so everyone will want to buy it. But ask who the single most influential entities in crypto are. It's not the blockchains — it's the exchanges. They hold enormous amounts of crypto themselves, they influence the media, and they aren't tied to any one chain. They make money when people are confused, when people buy coins and just leave them sitting on the exchange.
ICP breaks their business model
Here's the key. Every other blockchain needs a third party to access it. Even "not your keys, not your crypto" doesn't really apply, because whether it's a browser extension, an exchange, or a hardware wallet you could lose, you're still trusting someone to hold it on the blockchain for you. ICP changes that: you can hold your own crypto and interact with the blockchain directly, no exchange required. Since I got into ICP, I've had almost no use for exchanges beyond buying and selling ICP itself. Now multiply that across everyone — that's a direct threat to the exchanges' entire model. And it gets worse for them: if the other chains collapse in value as ICP becomes dominant, the exchanges stand to lose the most, because those coins have little value outside of being traded and held on an exchange.
The launch price that looked like a trap
This reframes ICP's launch for me. Watching Bobby O's analysis, the point that stuck was how the initial listing price was set. ICP listed on exchanges like Coinbase and Binance at hundreds of dollars — roughly 100x the few dollars of the last funding round — reportedly anchored to FTX futures, on an exchange that didn't even have spot trading. If there were any integrity to it, ICP should have launched around a few dollars and had an exciting run up. Instead it launched sky-high and crashed ~99%, which let every bot online call it a rug pull and blame DFINITY. If you were an exchange that saw this coin as a threat to your survival, crushing it at launch is exactly the self-interested move — not necessarily evil from their point of view, just people protecting their houses, cars, and business.
Why price obsession keeps it hidden
I've made roughly a thousand of these videos over the years, and the one thing people always come back to is the price. It can do AI on chain, it's a full-stack blockchain — "but what about the price?" If you can manipulate the price, most people close their eyes to everything else. It's like healthcare, where confusion keeps the system profitable: learn what actually keeps you healthy and that's "dangerous" information, so the antidote is constant noise and contradiction. The difference with ICP is that the technology is so good there are forces outside crypto that want it.
Where the hope is
Governments want secure sovereign clouds; cybercrime is being treated as a national security issue; tens of thousands of applications are being built on Caffeine AI per the recent DFINITY reports. Local computing is cheaper for casual work, but when you need things to be genuinely secure — as governments, companies, and businesses do — ICP is the best option I've found. The one thing we can do is educate people. As long as we fixate on price, nothing changes; when we focus on the technology, minds change. All my investments are ICP right now. If you want the deeper case, watch my ICP playlist, and I broke down the Bitcoin comparison in Bitcoin is a $1.6 trillion spreadsheet.
When Ethereum DeFi Collapses, Money Will Flow Into ICP
From my video published June 14, 2026.
Ethereum DeFi is a house of cards that could collapse at almost any time — and when it does, I think the money is going to flow into the Internet Computer. This is my prediction and opinion, not financial advice, and another creator has done an even more in-depth breakdown than I will here. But the short version is hard to unsee once you've seen it.
The recursive house of cards
Right now people are looping their positions: take Ethereum, get a liquid-staking token, dump that into another protocol, get another token, into another, and another. It's a snowball that can turn into an avalanche the moment one piece breaks. Everything is connected to everything else, and there are far too many centralized points of failure holding it together.
AI hacking is the trigger
All of it sits on infrastructure that's highly vulnerable to AI-powered hacking, and the obvious target is the enormous money parked in Ethereum DeFi. The cracks are already showing — there are situations where hundreds of millions of dollars have to be injected just to keep something from going to zero. It only takes one hacker pointing an AI tool at the infrastructure, finding the weakest link, and setting off a chain reaction.
When one thing fails, it all fails
That's the danger of the design: a single hack can cascade. One protocol fails, which breaks the next, liquid-staking tokens go to zero, and the prices that pumped on the way up dump on the way down — potentially dragging much of Ethereum and a pile of weaker coins with it. Honestly, it amazes me a little that it hasn't happened yet.
The NNS is vastly superior
Here's the alternative that makes DeFi look fragile by comparison: the Internet Computer's Network Nervous System. You can lock your ICP for two weeks at around 2.22% and unlock it cleanly, or lock for two years at roughly 6.65%. There are no third parties and no house-of-cards collapse scenario built into it — you're earning yield directly on chain on an asset I believe is appreciating. I broke down how I structure those locks in my ideal ICP accumulation strategy.
Real governance, not a black box
The other piece is governance. Your staked ICP isn't just earning — your votes actually control the network. DFINITY holds enormous voting power but generally steps back and lets the community decide the big proposals, rather than bullying the outcome. When I vote, my neuron moves around half a percent of all voting power. That's the best governance in crypto, and it's the opposite of the opaque, centralized machinery underneath Ethereum DeFi.
Where the money goes next
When Ethereum DeFi finally breaks, people are going to be desperate for a safer place to earn yield on an appreciating asset — and I don't think they'll find anything better than ICP. I don't want anyone to lose money, but destruction clears the path: collapses are when people finally learn, level up, and the whole space matures. So do your research now. Why centralize an Ethereum liquid-staking token through a chain of third parties when the NNS exists?
We'll see what happens — and none of this is financial advice. For the full case, watch my ICP playlist here.