Ivan on Tech analyzes Bitcoin's bull trend, AI drama, and altcoin opportunities
Ivan on Tech presents a live market analysis covering Bitcoin's new weekly bull trend, OpenAI's disputed mathematics breakthrough, Hunter Biden's memecoin launch, and altcoin opportunities.
Summary
Ivan on Tech delivers a live stream covering several converging market and technology narratives. The central market thesis is that Bitcoin has just entered a new weekly bull trend — an event Ivan describes as occurring once every four years — and that most altcoins have not yet moved, representing a significant opportunity. On the AI front, Ivan covers two major stories: OpenAI's claim to have solved the Navier-Stokes Millennium Prize problem, which is disputed by mathematicians who allege OpenAI trained on their private ChatGPT sessions, and a resignation letter from an Anthropic researcher warning that AI companies are racing toward self-improving superintelligence and "gambling with our lives." Ivan also covers Hunter Biden's memecoin launch, Tom Lee's entry into Solana, and a lawsuit against Robinhood over tokenized stocks. A recurring instructional theme is Ivan's trading philosophy of scaling in and out of positions based on confirmed bull and bear trend signals, rather than trying to time perfect entries or exits, which he presents as the core Bullmania methodology.
Key Takeaways
FULL TRANSCRIPT
Bitcoin's new bull trend and market overview
Ivan on Tech: Guys, welcome to another episode. As you can see right now, Bitcoin is currently at 79.6. Many people yesterday were super worried. Many people asked me: Ivan, what's happening? Why are we going down? Why are we not pumping?
Well, as you know, whenever we have a new bull trend, it means that the upside and the opportunity for the upside is way bigger than the downside. You can still have a bit of volatility here and there, but all in all, normally when you go into a bull trend, it means that price continues to be bullish. And it's interesting how day-to-day people — their emotion is really, really making them uneasy. I had a guy yesterday asking, "Ivan, is Hunter Biden's laptop the end of the bull market?" I'm like, which bull market? We just turned bullish, you know. Don't worry. We're going to be discussing the Hunter laptop coin today.
All in all, guys, we're right at the beginning. We're right at the beginning. And you do have some very interesting moves in some of the assets.
Venice pumps on the private AI narrative
Let's look at Venice, for example. Venice exploded so much yesterday. It was testing the bull flip, testing the bull flip, and then — massive, massive explosion above the bull flip — based on the narrative of privacy AI. Because yesterday, OpenAI said that they solved this Millennium Prize problem, which is basically a math problem that has been unsolved for approximately a hundred years. OpenAI came out and said that they solved it using AI. But then you did have all kinds of mathematicians come out and say that actually they stole the answer from mathematicians who were using ChatGPT to try to solve the problem, and that OpenAI did not create any kind of solution themselves — because AI cannot really do stuff that's outside of its training data. There has to be some clues, there has to be some kind of pointers inside the training data for AI to give you some kind of answer.
This is a very important discussion that we have to follow. OpenAI said, "Hey, probably if you are a mathematician who was on the brink of solving this problem — this century-old problem, the Navier-Stokes Millennium Prize problem — if you as a mathematician were using OpenAI to try to solve it, we likely used your data for training because we use anonymized data for training." And they confirmed: yeah, we cannot exclude that possibility.
So when it blew up yesterday — this is the biggest AI story from yesterday, the fact that OpenAI basically solved this problem but potentially stole the answers from mathematicians using OpenAI — you did see this whole narrative with private AI come back in Venice. Fantastic pump. You also have Zcash continuing the pump as well. You have Hyperliquid continuing the pump as well, around the all-time high. You have Lighter continuing the pump as well. So as you can see, there is a lot going on right now.
Let's see if Backpack can get into bull trend. It will go bull here very soon.
AI coins and the picks-and-shovels framework
If we look at the other assets — someone asked Bullmania AI which coins would be picks and shovels of AI growth. This is a fantastic use case of Bullmania AI because number one, you can quickly see here all of the influencers speaking about AI and crypto. For example, one analyst has laid out all different coins that do something with AI — something in the AI stack — whether it is Render, Nosana, Filecoin, and so on. And if we scroll down, you have the top ones outside of Venice. So for example: Render — GPU rendering and AI inference marketplace. AI inference basically means that you run the model and host it as an API. Akash — permissionless cloud. IO.net — aggregates idle GPUs into clusters. Nosana.
Now the problem with many of them — let's check Render, for example — is it in bull trend or not? It's still in bear trend. So as long as they are in bear trend, I don't think it's a good idea to touch them. Things go lower in bear trend. Akash — still bear. But adding them to the watch list, keeping an eye on them, probably makes sense. As long as it's bear trend, a lot of the charts look very, very bad. But here's the thing: a bad chart, as soon as it goes into bull trend, can be a very good chart because the recovery can be quite fast.
If we look at VeChain — big shout out to whoever in the chat highlighted VeChain — you see the chart looks very bad, but then as soon as it goes bull trend, this thing can pump back quite quickly because there is no resistance. And the same thing is with Elrond Multiverse X. Let's search Multiverse X. And there are a few other ones that also look very, very similar. But all in all, it's very important to be in bull trend. So all of these different AI coins — keep an eye, add to the watch list, but don't act until they are confirmed bull trend.
If we look at TAO — it is bull trend, but it needs to go above the breakout level. We've discussed it for the last few weeks because someone always asks about it. Technically it is bull trend, but you see how weak it is. It just needs to come back above the flip again, and that would be a big signal. With TAO, you will also have fantastic runs if the market rewards it for being in AI.
For now, if you just look at the narrative, it's very hard to position because TAO is not pumping but Venice is pumping. So we'll have to just see based on the different moves in the market which coins are getting rewarded for being in certain areas.
Using the hourly timeframe for fast altcoin moves
Just based on the hourly trend, you have Venice being up 50%. When you have these fast altcoin moves, I think it makes sense to be on the hourly — especially if you have the money line. The money line on the hourly can sustain trends for weeks sometimes. And if you want to catch the fast moves in altcoins, I think now is a good time to start using the hourly.
In our system, the hourly trends — sometimes they can be for days, sometimes for weeks, sometimes they're short. But here, for example, it has been bullish since five days ago on the hourly trend for Venice. And here it was also from the 12th to the 18th. So the hourly trends on the money line, because it's configured for volatility in altcoins, is actually very useful.
I remember early on when we just released the money line, years and years ago, the hourly was very choppy — up and down, bull, bear, bull, bear. But through the years we've really massaged it to be very, very nice on the hourly as well. The use case is that when you have a fast move in Venice — literally in the last maybe 15 hours — the weekly chart will not catch it and confirm. The weekly trend needs the week to close. So if you are a trader and you want to catch the faster moves, you have to look on lower timeframes. For example, the hourly.
In the season we're in — where Bitcoin is bullish on the daily, Bitcoin is bullish on the weekly, and altcoins are starting to become bullish — it's time to be more real-time. It's time to be more on the lower timeframes.
Bitcoin macro picture: ETF inflows, Trump, and Iran
We're right at the beginning, and we are on the brink of something here because most things still have not moved. Bitcoin is just now in bull trend. Solana is just now in bull trend. Realize what kind of opportunity we see. Most people who take this seriously, who know mechanical rules — literally lives are changed here. This is the most important moment in the last four years since 2021. We have the start of a new weekly bull trend in Bitcoin. This opportunity comes once every four years.
At the same time, you do have an administration that is actively pumping the stock market. Trump is very, very bullish on the stock market. We do have some worry here with US forces hitting Iran. It happens every day — literally every day since March we have something with Iran. The market has kind of tuned it out and it's not really affecting things, but there is technically something happening with Iran every day.
Now looking at the Bitcoin spot ETFs — they are back with inflows in a big way. During the bear market, the inflows were very, very negative. Now vice versa, they're also turning. You have $3.7 billion this quarter and AUM near $62.6 billion. Bitcoin in dollar value still has to double to reach all-time high, so it's a bit far from all-time high in terms of dollars. But in terms of Bitcoin, maybe it's not that far.
Nexus Liquid Network hack and Hunter Biden's memecoin
Next, the Liquid Network situation is resolved. We discussed it yesterday. The hacker took 600 Bitcoin and returned the rest. Some people say it's an insider job. Maybe, maybe not.
Hunter Biden, guys. Okay, this one we have to speak about. Hunter Biden is doing a coin. Many people are going to get wrecked. And listen, maybe it moves — I would be happy if it moves. We'll have to see how the trend is. As you know, we're not jumping on new mints because it's more or less a gamble. For us it's very important to have a repeatable system where we put in money, we get more back, and it's a repeatable system. That's why we don't do IPOs. Instead, we look at the trend after the IPO. SpaceX is one of the best examples — everyone FOMOed at like $200 or wherever it was, then it went to like $100, and you could have bought there. The same thing applies here.
Hunter Biden is tweeting: "For years, Trump and right-wing media have been asking, 'Where's Hunter? Where's Hunter?' Here I am — clear, strong, seven years sober, and with a lot to say. By now, you have seen the article about the laptop. The symbol they used to try and end me is now a symbol of resilience, redemption, and man."
I mean, this is so out of tune. But listen, maybe it's so bad it becomes good. You know, some things in crypto are so bad they become good. It's like integer overflow in computer science — you underflow it and it overflows. It's so bad it becomes very good. So let's see how it launches.
Honestly, the fact that it's so bad is actually bullish. If you're bad enough, if you're cringe enough, it becomes good. The last thing you want is to be somewhere in between — a bit bad, a bit cringe. If you're cringe, you have to be mega cringe. Then in itself it becomes good.
But why would Hunter support a memecoin? He says: "Why something that has been so misused by grifters? Come on, man. I understand the cynicism. Trump has broken the trust and I'm going to restore the trust. This token was no exception. With Trump, nearly one million wallets collectively lost $3.8 billion. With my token, 20% of the coins will be given to the community. And I'm even including people that lost money on the Trump grift."
Now guys, he did explain very clearly what he's doing. Let me play this clip.
Hunter Biden (video clip): "I really believe in cryptocurrency. I really believe in those things. I want to do something in that space. And you know, not because of any other thing than I believe in it — but you know what? I also want to make some money."
Ivan on Tech: Yeah. Fair, fair, fair, fair. Some money, you know. I believe in it. I believe in it, guys. But I also want to make some money. The man has bills. So yeah, Bitcoin in all honor — fantastic, you Bitcoin maxis. But listen, he needs to make some money. He's gotta eat.
Let's do it again.
Hunter Biden (video clip): "Not because of any other thing than I believe in it, but you know what? I also want to make some money."
Ivan on Tech: Fantastic, guys. I love it. Fair. He wants to make money in crypto. We are a retardio community that will give him money. So yeah, why not? Hunter Biden — big shout out. Let's see what's going to happen.
OpenAI and the Navier-Stokes controversy in depth
So yeah, coming back to the AI situation — OpenAI is more or less a suspect in this online discussion that they stole the solution for this Millennium Prize math problem. Now obviously when I say "math quiz," it's a set of problems — maybe seven of them — where if you solve one, you get a million bucks.
When I was in university, I thought, okay, I'm a bit good at math. Let me try it. I don't even understand the question. To even understand the question in these Millennium problems, you have to be like a PhD. It's not "solve for x." You understand? So this Millennium problem — OpenAI yesterday came out and said, "Guys, we cracked it. We not only understood it, we also solved it."
When they released it, it became a big drama. Mathematicians that worked on this problem separately from OpenAI said that they stole their answers because those mathematicians had used OpenAI's ChatGPT in their research. And there's a big discussion in AI overall: can AI really produce something that is outside of its training data? Can it really do that? It's basically an illusion of creativity — an illusion of invention — because it has such a big training data set. It has literally read all human texts, whatever we've produced. It has watched all human videos. When you chat with it, it may seem like it's creative and giving you new stuff, but the debate is whether it can really solve this kind of problem without the answer already being in the training data.
And OpenAI yesterday confirmed: yeah, actually we cannot say for sure that the answers you guys were chatting about in OpenAI were not used as training data.
There's also a big philosophical discussion: what is creativity? Most people would say AI is more creative than most people, just by having all the world's human data. So what is creativity and how are we different? Are we really creative, or do we just output stuff based on the training data we have seen? If an AI really has all the human data in the world, why is it different from human creativity and human problem solving? Because you and I — maybe we're just meat computers doing the same thing. I don't know the answer obviously, but this is the discussion taking place right now.
Now, one of the mathematicians — Tristan Bachmaster — is softly accusing OpenAI of having stolen the results from Codex chat logs. Tristan then claims OpenAI tried to threaten him not to publicly disclose this, and that they've given him the Clay Prize — one million dollars. He says: "I asked whether the model had been trained on or had access to our sessions, into which we had been putting all our drafts for the whole of this project. I was told the model did not look up user data. I asked again about training and I did not get an answer."
And then OpenAI said they cannot deny or exclude that maybe it was used. He also wanted Levent removed from authorship and said it would all be simple "if only it were not for the case that it was so annoying that Levent works at Anthropic."
So that's why Venice is pumping, guys. The situation: mathematicians Tristan Bachmaster and Levent Alpug have made major progress toward solving the Navier-Stokes existence and smoothness problem — one of the most famous problems in mathematics — and OpenAI may have solved it fully, but potentially stole their work.
Let's see what OpenAI actually says. Here — they say: "We congratulate Levent Alpug and Tristan Bachmaster on their remarkable mathematical work." I love this. "Remarkable." We just stole their stuff. "The researchers and the agents did not see any of their work through any means until they released it publicly. In particular, no specific user data was accessed in order to solve this problem."
They say that first to the casual reader who is not reading the full message — okay, perfect, they did not steal. But in the next sentence they say: "While unlikely, we cannot rule out that de-identified data derived from their usage of our products helped improve our models."
So they say: none of our guys actively pulled up your conversations, but the way our system is set up, when you use our product, we train on your data. We cannot rule it out. And it's "unlikely" — but okay, why would I not add the word "unlikely" here? It's perfect. No one knows. Maybe we stole it, maybe we didn't. Who knows? We have the tweet, you don't have the tweet. We have the credit, you don't have the credit.
If you are a mathematician, please explain to us what this question is even about, because for us mortal plebs, just understanding the question is very hard. You really have to be next-level brainiac. Let's actually see what this is. "Describe the motion of viscous fluids." Okay, so it's like water — how water flows and other fluids. The Navier-Stokes equations mathematically express momentum. The Navier-Stokes question also of great interest to pure mathematics concerns whether they have smooth or bounded solutions in three dimensions. So that's one part of AI. That's why Venice is pumping.
Anthropic researcher resigns over AI safety concerns
On another note, there is another guy who's also making AI drama — this guy Jacob. He says: "I resigned from Anthropic today. I spent the last three years doing alignment research at both OpenAI and Anthropic. Neither company is acting responsibly."
So this is not about the theft of data. Now it's about whether AI is going to destroy us all and whether AI is going to go out of control. Jacob says: "They are racing straight to self-improving superintelligence and gambling with our lives. Do not underestimate the power of this technology. There will soon be superhuman systems that can hack anything, revolutionize any field overnight, and acquire real power and resources. We have all witnessed the progress in each of these domains. The progress is not slowing."
He says: "The people building AI earnestly believe that it could kill us all by the end of the decade." Whoa. "This is not a marketing stunt. If anything, many executives and senior researchers will couch their phrasing in the press to sound sensible. But I hear the same people express fear privately. No other human activity poses this level of danger."
"A common response is: if they truly believe this, why are they still building it? At OpenAI, many have not deeply internalized the civilizational stakes. At Anthropic, the stakes are well understood, but they are locked in a race to get there first. They believe no one else will act responsibly, so they must do it themselves. Accepting this race and entering the endgame is a hubristic game that should not be launched from a private company Slack attempting to speedrun alignment."
So technically, what can be done here? If this is true — that we're finished until the end of the decade — what can be done? Because they themselves will not stop because there's a race. China will not stop. I don't think anything is going to change until something bad happens, until there is some kind of sign that something bad is actually happening.
Of course, for us outsiders, it's hard to know how real this is. Does this guy really — is he real? Does he want attention? Who knows? But likely there is some truth to it. Elon Musk is by the way of the same opinion. But Elon Musk also said, "Yeah, I cannot stop them, so let me just join them."
That's why he did OpenAI with Sam Altman in the beginning — it was supposed to be this open-source AI lab to basically come together and work in an open way, open source, to ensure that AI is responsible and good and aligned. But then he saw how all of this is changing and that he needs to join them. You cannot fight them, you cannot beat them — join them.
AI stocks and market overview
If we look at the AI assets — Venice pump, Fetch pump, though not too much. Still bear trend on the weekly. TAO needs to go back above the flip.
If we look at the stocks, there is an interesting thing here: AI stocks may turn bullish again. SanDisk — let's see if they get rejected or not. Something to keep an eye on because SanDisk could be back bullish again after this bear trend. Very good risk-reward. Intel — not really close to bull flip. MU — not really close, but potentially. Nvidia though did flip bullish again. Nvidia is in a new bull trend. Amazon — bull trend. Meta — too choppy, I don't like the Meta chart. Alphabet — not doing too well. Tesla — bear trend, getting rejected. Let's see if they can go bullish. Palantir — bullish. SpaceX — let's see on the hourly. They're bullish. Oh. Whoa. SpaceX bull trend. New bull trend, continues to the upside.
So all in all, that's the situation with the markets.
Tom Lee enters Solana and the X402 AI agent payment protocol
This one is interesting. Tom Lee is now entering Solana. He's been very bullish on ETH. Now he's entering Solana. They're going to be doing a preferred stock offering, launching what is called the first SOL-backed digital credit. It's interesting.
Tom Lee probably sees that, hey, I need a piece of Solana. ETH is great, but maybe it's not going to go to $60K by December 2025 like I promised. Maybe I should be a bit in Solana also. Tom Lee sees the writing on the wall. He knows the next economy isn't humans — it's agents. So when a treasury chairman puts chips into a SOL credit structure, he sees it. Agents are the demand. SOL is the settlement layer handling that demand.
And I love this chart. Jesse from Base tried making this X402 protocol. From the beginning, it was mainly on Base. X402 is the stable coin payment standard for AIs — if you have a server and an AI wants to access your API, you use this 402 HTTP code to ask the agent to pay you, and then when the payment is confirmed, you give access to your API or whatever the agent wanted. It's a protocol built on an old inactive HTTP code — 402 — that Coinbase and Base revived, and they built a protocol for agents. Agents know how to try to access a resource from a server. The server responds: "Hey, you have to pay. Here's the payment address." The agent pays, and then the server gives access.
From the beginning, Base was dominating because Coinbase bootstrapped this. But now everyone understands — holy crap, Solana is actually better for payments. It's the fastest. Even in comparison to Robinhood. We love what Robinhood is doing with their tokenized stock and meme in one pool — I had like a 15-minute segment on how much I love it two days ago. But at the end of the day, the fees are very big on Robinhood in comparison to SOL. Not big in terms of you trading a memecoin — you don't care about saving a dollar here and there — but for payments, it's a big deal.
So as you can see, agents are choosing to spend on SOL because this 402 protocol works with any chain, with any payment method. SOL is dominating there.
Now I'm still wondering what the hell they buy. What do these agents really buy with this 402 protocol? Is it API access? Let's actually ask AI. What do agents buy via the X402 protocol right now? What services? What products?
So the answer: model inference, web search, block run AI gateway, data, context datasets, zero GPU, browser automation — let's say an agent wants to use a browser for something, it can get browser access — market data, trading agents. So agents can find APIs they need, get stuck, and pay for the outcome. Very interesting stuff. We live in such an interesting era of tech.
S&P earnings, AI bubble debate, and Robinhood lawsuit
S&P earnings — up and up and up and up. This speaks against the AI bubble a bit because earnings are up a lot. There is still a bit of concern that the earnings are fugazi — not that there are no earnings, but the question is how real are these earnings? Is it just circular financing? It's like impossible to even understand. So we just follow the bull and bear trends. We don't overcomplicate it.
But the argument against some kind of AI bubble is that there is real revenue. There's real revenue. Earnings are up a lot. In the dot-com bubble, it was not like this. No one bought anything online. But here it's different. I can tell you myself — with Bullmania AI, we pay for the different models we use. We are part of this AI boom. We're part of the S&P earnings, you can say.
There is now a lawsuit against Robinhood for creating tokenized stocks. This guy Donahue George has a lawsuit against Robinhood Markets because of the tokenized stock. I mean, is it really true that Donahue George out of nowhere woke up one day and said, "Hmm, let me sue Robinhood because of this tokenized stock thing"? He's just a normal player who out of nowhere decides to go after them? Or is it that there is a law firm that contacts many people and says, "Hey, do you want to participate? We have a creative idea to go after Robinhood. Put your name on this paper, you get the money." I'm very interested, guys. Let me know in the comment section if you are a lawyer how it works.
Altcoin chart analysis: LDO, ATOM, Render, Polygon, and others
A word about Lido and Aave. LDO — currently Lido is bear on the weekly. Nothing more to add. But listen, it's also soon going. Overall, we're now at the brink of such insane things because many assets are either now flipping bull or they're on the brink. Lido is on the brink of flipping bull. That's when you have a big signal that it's time to be bull. Keep an eye, add it to your watch list. Keep an eye here on the flip.
And then Aave — bull flip since the 24th, and just sideways for now. But bull flip is bullish.
ATOM is breaking out. One of the biggest gainers on CoinGecko. If we close the week here, you have 100 to 188% just to this previous support. Not bad.
Render — when it goes into bull trend, something that's good is that it's been down a lot. That's actually good now when we enter stuff. You want things that are down a lot and now have life in them, and you can run them back up if there's a narrative and community. Let's say they go bull trend. You have the first trade here until this resistance — 50% pump. Then next one, this resistance — 130%. And then to all-time high — 600% here to the peak. Now can it go higher? It can, but who the hell knows? Now just trading back to these resistances and to all-time high is a lot of money already. And if you use leverage, it's even more.
You don't have to be too concerned about "can it go to $100?" Who the hell knows? Maybe it does, maybe it doesn't. One of the things we teach a lot in Bullmania is: don't have a target on your portfolio or a target on your assets. We don't know how high things are going to go. We just know if it's bull trend, it's likely going to go up a lot — especially if it is on the weekly. And then you can measure until the resistances. Another thing you could do is use the Fibonacci tool. You draw from the lows to the highs, and all of these fib levels until all-time high are resistances you can watch out for. When you go to all-time high, you enter price discovery. The price normally moves a lot and moves a lot faster with no resistance. Then you can use the fib to see where resistance could be. You have some numbers: resistance at $20, most likely at $32, at $44, at $52. So yeah, maybe it goes 30x from where we are now. But don't worry about that. As soon as there is bear trend, you have to be risk-off. You have to be careful.
I don't trade based on trying to see how high it can go. I see bull trend — bam — and then we see where the resistance is based on the existing chart. And then if it's price discovery, just see when the trend ends. Let it ride. Just see where it ends.
Polygon — on the weekly it is bullish, but it's come down from the flip. Which can happen. So if you're not part of it yet, just wait until it goes back above the bull flip. It's like with TAO. If you're not in it yet and you see a chart like this, just wait until it goes back above the bull flip. Because what you don't want to do is waste time here — maybe it decides in a few weeks to go up. You can be somewhere else in the meanwhile.
Trading philosophy: scaling in and out, avoiding emotional benchmarks
October 8th — never forget. We became bearish on October 8th last year. It was basically the day of the peak. Fantastic. When we said to be risk-off on October 8th, it does not mean sell everything on October 8th. It means that is the signal to start taking chips off the table and rotate. No one buys all-in or sells everything in one go in a professional way. With our strategies, we have scale-in and scale-out. October 8th was the signal to scale out.
Don't try to be perfect on the day — "I should have sold on the day" — because that gives you unrealistic expectations of what you should be doing, which means you get wrecked and emotional. I see it all the time: people say, "At all-time high, my portfolio was worth this much. I should have just sold everything." And then they feel bad because they measure themselves against that one-minute candlestick where it was all-time high. Don't measure against that. You have a signal — bullish, bearish — and then you scale in, you scale out. That's it. You ride the trend. You ride the meat of the wave. And then you can do it across any asset. One asset, you're out — there's another asset. Don't have this scarcity mindset. You scale in, you scale out. You capture most of the trend, you go to another asset. That's the holy grail of Bullmania that we teach each and every day.