In a long-form interview organized by PANews, Cobo co-founder Disco Fish and Jilian Technology founder Jin Ming discussed AI, entrepreneurship, investing, Bitcoin, Web3, agents and questions of personal purpose. Disco Fish entered crypto in 2011, while Jin Ming has focused on video AI technology and commercial applications for years.
AI has cut the cost of turning ideas into reality
Disco Fish said one of the clearest changes brought by AI is that many tasks that once required heavy time and capital can now be completed quickly if someone simply has an idea. In his view, that shift makes human will more important.
He said that in the past, a person could have a seed of an idea or a dream and still never move beyond that stage because execution on the back end was too expensive. Now those costs are falling quickly, and AI has compressed and structured a large body of human knowledge, making those capabilities available at much lower cost. As a result, a much wider range of ideas can be delivered fast.
For him, AI has opened a much larger space of possibility. Things that would have been hard to imagine 10 or 20 years ago can now move from idea to delivery with far less friction.
He still sees himself as a builder, even without a clear map
Asked whether he is satisfied with his current life, Disco Fish said things are “okay,” though he still feels some uncertainty. He contrasted that with his early startup years, when his sense of mission was much stronger.
Now, he said, he is in a more chaotic phase. AI has expanded what seems possible, and he sees this as the best of times in that sense, but he has not yet found a path that feels fully clear, either in terms of mission or in terms of a concrete business route. For now, he is still exploring and working on improving himself.
Looking back on his early years as a founder, he said that when a real window opens, it creates an intense creative impulse. It feels like the kind of thing that would become a lifelong regret if left undone, and that force pushes a person into building it.
In 2013, the opening was mining infrastructure
Disco Fish said the opportunity he saw early on was not just “mining” in the abstract. In 2013, mining had first taken hold in the US and then began shifting to China. At that time, early ASIC mining machines in China still had to connect to US mining pools.
That created two practical problems. The first was network performance: latency and packet loss on the international internet hurt mining efficiency. The second was that settlement across the mining industry was not structured well at the time. He said the work they took on was to solve those two issues.
He described this kind of startup impulse as the result of spotting an unfilled niche. When a market has unmet needs and you happen to be able to address them, the urge to build can become very strong.
“Being alive means adding a little more possibility to the world”
On purpose, Disco Fish said he still wants to make things and continues to define himself as a builder. Since the arrival of the AI era, he has built various small tools and dashboards, but he has not yet found the one large, systematic project he wants to commit to. He is still watching closely.
He also said that after spending years on the front line of entrepreneurship, confusion can deepen rather than fade. That forces people to go back and study history, philosophy and older patterns of thought, and those ideas can in turn shape present decisions.
When Jin Ming framed life as a balance between leaving a legacy and seeking experience or happiness, Disco Fish said he does not strongly identify with either extreme. He does not feel driven to leave something behind for its own sake, nor is he focused on chasing experience. What he feels instead is an internal urge to build. Making something real gives him satisfaction. As he put it, being alive means that your existence makes the world’s set of possibilities slightly larger.
Anxiety exists, but he manages it by switching modes
Jin Ming noted that many people who want to be builders also tend to be anxious, because if your life purpose is to build, you can feel uneasy whenever you are not moving actively forward.
Disco Fish said that is true “to some extent,” but added that he has learned to switch states. Since childhood, he has had ways to empty his mind, including caring for animals and spending time on hobbies. He said he generally enters that kind of mental reset every week and sees it as necessary.
Freediving, fear and death anxiety
The conversation then moved into freediving and fear. Jin Ming said freediving is almost like a spiritual practice. While holding one’s breath, the body keeps sending signals, creating a confrontation between reason and instinct. The process requires repeated positive self-suggestion, telling yourself that you are safe and entering a meditative, flow-like state.
He explained that on land, it is difficult for a person to actually suffocate themselves by holding their breath. At the limit, the brain may briefly black out, or BO, entering a short shutdown. About 10 seconds later, the airway reopens and breathing resumes naturally. Underwater, the danger is different: if a person loses consciousness there, the reopened airway may take in water rather than air. That is why freediving training must involve a buddy.
Jin Ming said that beginners usually start well below their physical limit, often around 50%. After 10 sessions, or two to three months of training, they may gradually move toward 70%, 80% or even 90%. In his account, what stops newcomers is often not the body itself but a lack of familiarity with the body. As trust grows, the process can become far more peaceful than people expect.
On death, Jin Ming said long-term freediving and breath-hold training sharply reduced his fear. He said he felt fear in the first few sessions, but no longer does. Disco Fish replied that he still has anxiety at some level and said even his “agent companions” would tell him that death anxiety remains part of his inner state.
AI agents are now his main information gateway
Asked how he gathers information every day, Disco Fish said, “almost entirely through AI.” He said that when he wakes up, AI tells him it has already read everything he engaged with the day before.
He said he currently uses three AI agents. One is highly instrumental and tells him what he should do. Another has a softer, more feminine character and tries to surface deeper emotional material. The third reviews what he did the previous day and points out blind spots it has noticed.
He gave an example in which an agent might tell him that a specific chapter in a historical book discusses a question that closely matches his current situation, then ask whether he wants to spend 15 minutes reading it. The materials would then appear immediately for review.
He also maintains a note-taking system that links his past interests with things connected to his daily life through what he described as an evolutionary lens. He compared his own attention to a kind of “god” that decides which ideas or propositions survive. AI then “crossbreeds” new combinations each day, filters them and recommends what to read.
Because of that workflow, he said he rarely reads material directly anymore. Instead, he consumes AI-filtered daily digests and longer essays, lets AI process them first and then compresses the output into PPT format for rapid review.
He cares less about immediacy and more about long-duration decisions
On whether AI mediation could leave him behind on breaking developments, Disco Fish said he does not care much. The most important information, he said, will find its way to you anyway. He also said he does not trade on a very short-term basis.
He added that he realized early on he had no special talent for short-term trading. Simple decisions come easily enough, but he sees himself as better suited to decisions with a longer time horizon.
His real investment framework formed after DeFi
Disco Fish said he only truly started investing after going through DeFi. In his telling, DeFi forced him to replay the history of investing and finance, from the “underground Federal Reserve” idea all the way through how traditional finance was built over roughly 200 years.
That process raised many questions, which pushed him into studying history and financial theory. Only at that point, he said, did he begin to invest seriously. Before then, he did not have a complete investment framework.
Looking back at his years in mining, he said he did not even consider what he was doing to be investing. At most, he had an instinct that he should hold some Bitcoin, along with a rough awareness of position management.
He also said he is an ISTJ and tends to abstract lessons out of practice. He forms some frameworks first, then trains his intuition, runs into new questions and abstracts further. That is how his investment framework developed over time.
From DeFi to RWA, he focused on what products actually do
When Jin Ming noted that very few people develop an investment framework through DeFi because much of DeFi often looks like “farm, sell, withdraw,” Disco Fish said the key was not that surface behavior. What mattered was understanding what financial functions those products provided and what market demands they were serving.
He said DeFi was building a financial market on-chain from 0 to 1. That was the process he paid attention to, not just arbitrage or yield farming. The experience created intuition and “muscle memory,” but it also generated a large number of questions. Once those questions began to make sense, history started to connect.
He said the current wave of RWA on-chain and the introduction of US equities onto blockchains has extended that understanding into traditional markets. In his words, it puts US stocks right in front of you and forces you to understand the logic behind them.
He added that when people used to ask him for money to trade US stocks years ago, he did not really understand the point. Later, a growing pile of questions, combined with AI-assisted learning, let him move through books and materials much faster and revisit why those market structures exist.
“A good asset does not need a narrative”
On Bitcoin, Disco Fish said his view has not changed. BTC remains digital gold in his framework. He said very few assets can be held without intermediaries, while many others sit behind one, two or even three layers of intermediation.
As for the market’s pessimistic voices around BTC, he said Bitcoin’s value tends to reveal itself under extreme conditions. For that reason, he said it is natural for many people to ignore it when markets are thriving.
Responding to the idea that Bitcoin strengthens the sovereign individual in the tradition of classical liberal thought, he said one core trait of the sovereign individual is control over one’s own assets. AI strengthens personal capability, while BTC is a core asset in how a sovereign individual can manage and dispose of wealth. He sees both as important.
When asked about people who were deeply committed to BTC early on but now say they no longer see a fresh narrative, his answer was blunt: good assets do not need narratives. In his view, time itself becomes part of the proof.
Agents may be the next key test for Web3
Jin Ming said he sees only two major Web3 narratives in the current cycle: Ethena’s all-asset arbitrage model and RWA on-chain. But in his view, both are still early and far from fully mature.
He added that the real breakout consumer application in Web3 has not yet appeared. He was not referring to products used by investors or traders, but to something ordinary users could adopt the way they use WeChat, changing everyday life at scale.
Disco Fish responded by repeating a view he says he has held for a long time: ecosystems such as Ethereum may not end up being built primarily for human users. They may be built for agents. In that structure, the chain becomes a network and humans may not be the most suitable end users.
He said many on-chain products still demand too much technical competence and security awareness from individuals. If agents become the middle layer, the interface to those systems could be much better.
Asked about Ethereum and Solana, he said the question he is watching most closely is whether agents can achieve large-scale adoption in an ecosystem like Ethereum. If that happens, he said, it could mark a turning point.
Web3’s underappreciated feature is shared economic ownership
Jin Ming argued that the most underestimated part of Web3 is tied to the progression from “read” to “write” to “own.” He said that shift may change how wealth is distributed across society.
He used China’s A-share market as an example. Investors trade through software, but a large share of the resulting fee income flows to platforms such as East Money and Tonghuashun, not to the users themselves.
In Web3, by contrast, participants can also become economic owners. He cited Binance and Hyperliquid as examples where users generate fees for the platform through trading, while ecosystem participants may also share in platform growth through tokens, buybacks or similar mechanisms. In that arrangement, a user can be both a participant and, in effect, a shareholder.
Jin Ming said that may be one of Web3’s central values, and he argued it should not remain limited to on-chain settings. In theory, any money-related field in the future could adopt a similar structure, where participation, financial return and exposure to ecosystem upside come together.
Disco Fish said he believes similar systems will appear over time. Data in the AI era, he said, may itself support that kind of mechanism. The possibility of technological diffusion is already open, though large-scale adoption will still require time and the resolution of many problems. He added that successful demos already exist in some edge cases, even if they have not scaled broadly.
Gold or BTC? He prefers concentration over diversification
On gold, Disco Fish said he does not pay much attention to it and stated directly that Bitcoin is the better gold.
Jin Ming pushed back by saying the two assets share a common base in consensus, though not among the same groups. Bitcoin has built part of its consensus around sovereign individuals, while gold has stronger consensus among sovereign states, especially central banks, which continue to buy it.
Disco Fish framed the split as two camps: “digital currency plus AI” on one side, represented by Bitcoin, and “gold plus industrial manufacturing power” on the other. He said his own orientation sits more on the former side.
When Jin Ming argued for holding both Bitcoin and gold to benefit from lower or even negative correlation, Disco Fish said he does not really believe truly negative-correlation assets exist in the strong sense people often imagine. Many assets may look diversified under normal conditions, but they still share a small number of core constraints. When those variables move in a systemic way and markets enter extreme conditions, many assets end up behaving similarly.
Even when Jin Ming cited examples such as oil against some credit assets, or US Treasuries against risk assets, Disco Fish answered that there are also periods when Treasuries and equities fall together. In his view, the real problem is surviving extreme cases. An apparent hedge may not protect you when stress becomes most severe.
That is why he does not see himself as a classic diversification advocate. He prefers concentration, because no one can truly understand everything and attention is limited. If the underlying logic of an asset is not fully clear, he would rather focus capital on a small number of things he really understands.
Investing comes down to temperament, conviction and position sizing
Jin Ming said some assets may not require complete understanding. Gold, for example, may still deserve a place in a diversified portfolio because of the long-term consensus around it among central banks.
Disco Fish replied that if an asset’s drawdowns disturb your psychology, the simplest solution is not to hold it. By contrast, he said Bitcoin declines do not create much panic for him, but that only works if the holder has genuine conviction. And he added that such conviction requires a lot of preparatory work.
Jin Ming said the bigger challenge for ordinary people is that most are not professional investors. Position sizing, drawdown control and emotional discipline are all difficult to manage well.
The conversation then returned to training and personality. Jin Ming said athletic ability is partly trained and partly innate. He expects more people in the future to treat sport as a necessity, invest deeply in a discipline and gradually become “athletes” in a meaningful sense. That process can bring both satisfaction and greater happiness.
In his view, all of that adds up to a more complete trading lifestyle. With the right broader lifestyle in place, Bitcoin drawdowns can look like opportunities for Class A investors rather than sources of distress.
He offered a concrete example. At $120,000, he said, Bitcoin feels hard to buy. Above $100,000, he would not be especially eager to accumulate either because the market has not given him the chance he wants. Below $90,000, however, he would see opportunity. He said he would buy gradually at $90,000, $80,000, $70,000 and $60,000 because his long-term view remains positive.
Is AI cyclical, or the next era’s utility layer?
Jin Ming said he does not think it is accurate to describe AI as a purely cyclical equity theme. His view is that AI will become the most important productive force of the next era and replace a large share of the systems the current world relies on. If that assumption is right, then AI is not just another cycle. It is the new era’s “water, electricity and coal.”
Disco Fish responded with a historical comparison. During the 2000 internet bubble, he said, people were also convinced the internet was the future. At the time, network infrastructure was viewed as the core opportunity, and many people bought Cisco. Cisco then spent a long period working through valuation compression. He asked whether AI could produce something similar.
Jin Ming said anything is possible, but argued that if the AI era truly arrives, storage demand will become enormous. Today’s mainstream AI products are still mostly conversational agents and some coding agents. The much larger demand, he said, may come from embodied intelligence — actual robots.
He noted that homes still do not have true embodied robots in any broad sense, and workplace robotics remains concentrated in a relatively small number of factories. Large parts of white-collar work have not been meaningfully replaced. Beyond that, he said, the human brain itself may eventually require more external storage, whether through multiple backups or through brain-computer interfaces that create an “external brain.” Those demands do not yet exist at scale, but if the AI era arrives, the addressable need could be huge.
For Jin Ming, the real question comes back to whether one believes the AI era will arrive. His answer is yes. He also said robots are no longer a completely wild assumption. Following the present trajectory of technology, he believes the possibility of massive production is already visible.
How he filters for core assets
Disco Fish said that this ultimately loops back to trading lifestyle. A person’s temperament shapes the investment philosophy and framework they choose, which then shapes position management and emotional management.
Jin Ming said people differ widely, but those who do well over time usually develop an internal system that is distinctly their own. Even the definition of a top-tier asset varies from one framework to another.
Disco Fish then laid out his process in three layers. First, he looks for an anomaly — a company that seems unusual in some way. He does not open with a heavy position. Instead, he may buy no more than 2%, and often only a few tenths of a percent, as an observation position.
From there, he spends time studying it. If the logic checks out, he may upgrade the position from a tiny weight to a mid-single-digit portfolio allocation, but usually not above 10%.
Moving from roughly 10% to around 20% as a core holding requires what he called a leap of faith. At that stage, fundamentals and analytical logic are no longer enough on their own. He wants to see a sufficiently large vision, a large total addressable market, sound business logic and a genuine strategic choke point.
That leads to the framework he currently sees as most central: “monopoly plus growth.” In his view, any asset that deserves the core tier must have a very large TAM and a sufficiently large vision behind it. Without that, it does not belong in the most important part of the portfolio.
Current core holdings: Bitcoin, Ethereum and Tesla
Asked which assets currently make that cut, Disco Fish named Bitcoin, Ethereum and Tesla. SpaceX, he said, is still in a market-timing and observation phase and has not yet been upgraded into that core set.
Ethereum remains inside the group for now, though he said they have recently been discussing whether it should be moved down from that tier.
Jin Ming asked whether identity also plays a role and mentioned that when Disco Fish sold his ETH a year earlier, he had said in a group chat that he was “sleeping better,” to which Disco Fish appeared to agree at the time.
Disco Fish said that was true because he was no longer mining DeFi. Still, he said the main logic he sees today is that agents may eventually run on Ethereum, and that thesis has not yet been disproven.
When Jin Ming asked why agents could not run on Base or Solana instead, Disco Fish said they could. But he thinks Ethereum may be a better match because of the decentralized infrastructure still present at the base layer. He added that while many of those pieces have not fully materialized yet, the foundation has been trying to move in that direction, including work on agent registries and related ecosystem infrastructure. Nothing especially large has broken out yet, he said, but there has at least been visible effort, and that has left the market with some hope.