Editor | Wu Says Blockchain

This interview delves into the investment trends, market status, and future opportunities of the AI Agent project. Analyst defioasis shared his investment experiences in on-chain assets, memecoins, and AI-related projects, including the rise of key projects (such as GOAT, WorldCoin, Turbo, Pippin, etc.) and the market logic behind them. Additionally, it analyzed investment methodologies, position management strategies, signs of market bubble cooling, as well as changes in industry trends and the importance of project narratives.

Please briefly introduce yourself, and review when you started researching AI Agents, when you officially started investing, and what your overall return rate is currently.

defioasis: Hello everyone, I am defioasis. I am generally interested in on-chain data, on-chain assets, and the derived gameplay. Since last year, my main focus has been on on-chain activities.

In fact, my research on AI Agents also stemmed from on-chain exploration. I have been observing on-chain assets since last year, but I am not particularly skilled in pure memecoins; I mainly just watch and have not been able to convince myself to invest large amounts.

The turning point was mainly at the end of October and the beginning of November last year when GOAT skyrocketed to a market cap of several hundred million dollars in a very short time, and the listing of GOAT futures on Binance made me rethink this sector. Of course, before discussing that, I want to mention some AI-related events that occurred before GOAT. There were two significant events: one was WorldCoin, which was co-founded by OpenAI founder Sam, and WLD was treated as OpenAI's meme in crypto, with an FDV exceeding $100 billion at one point; the second was Turbo, which claimed to be an AI meme created by GPT, and it also surged 200-300 times in CEX last year. These two events show that people are still very interested in and willing to speculate on the AI + Crypto theme.

Now back to the end of October last year, after GOAT was listed on Binance futures, I quickly associated it with WorldCoin and Turbo, and Binance has the ability to lead the sector. At that time, there were four assets that I found very eye-catching, two of which were AI-related: one was ai16z, and the other was ACT; the other two non-AI-related ones were LUCE and BAN. Looking back now, ai16z seems to be much stronger than ACT, but at that time, the founder of ai16z, Shaw, was still a nobody, and ai16z was launched using daosfun, with some issues in the token contract minting, and the volatility was particularly high due to a small pool. I built a position in ai16z at a market cap of $25 million, spending about $3,000, but the pool was small and at one point it was fudged down to below a $10 million market cap, so I didn't dare to increase my position. Therefore, I ultimately chose to heavily invest in the more stable ACT, which had a larger number of holders. ACT can be considered my first official investment in AI. My first purchase was on November 5, where I bought $3,000 worth of ACT, and later I found it on Bitget, gradually purchasing over $10,000 worth of ACT at a market cap of $22 million, with an average price of $0.022. At that time, I didn't expect that I would continue to buy and hadn't bought enough, and then unexpectedly it was listed on Binance. I remember I was attending a conference in Bangkok and was quite shocked. After being listed on Binance, it reached a market cap of $700-800 million at its peak, but I actually sold off most of my holdings the morning after the Binance news when it was at a $500 million market cap; I haven't sold the rest since then.

Later, I also observed that AI Agents were the only ones transitioning from general PumpFun to verticalization and forming scale, so I delved deeper into this sector for investment. After ACT, I also identified some other good targets, such as Pippin, which I heavily invested in and achieved over ten times returns. The overall position for investing in AI has reached 7-8 times since November, and recently it has pulled back a bit, but it's still around 5 times.

What is your method for researching AI Agent projects? Can you provide an example of the AI Agent project you researched the most deeply to help others understand your research methodology?

defioasis: The current AI Agents or some AI-related assets are quite different from previous AI projects. Now, most are assets formed based on fair launches similar to PumpFun, so the project parties or founders may not hold much of the assets, possibly even less than some sniper or large holders. This means that the character, philosophy, and background of the founder are very important; otherwise, if they lack ethics or are not real-name verified, they could easily abandon the project and start a new one. Therefore, my methodology for AI Agents starts with the person: whether they truly want to do it, whether they have the ability to do it, and what they can achieve.

For example, let’s take the experience of the Pippin project. I first learned about Yohei Nakajima from the Solana AI Hackathon judges on December 11. The Solana AI Hackathon was a hackathon event organized by ai16z and Solana. When I saw the list of judges, I noticed a Japanese person, Yohei Nakajima, who is the founder of Pippin. I found it quite interesting that he was working on a child-oriented AI Agent, as I hadn't seen similar agents before. Additionally, being a judge rather than a participant in the competition clearly indicates a higher level of credibility. Further research on Yohei Nakajima revealed that he is the founder of BabyAGI, which has over 20,000 stars on GitHub. I found some information indicating that BabyAGI is quite impressive and has been cited by many media papers as a concept product of AGI, proving its strength. Furthermore, Yohei Nakajima is a partner at Untapped Capital, with major Web3 projects like Pixel, which has been listed on Binance. Overall, the founder of Pippin, Yohei Nakajima, has strong technical and capital resources, and as a real-name verified individual with a certain level of prestige, the probability of him rug-pulling is much lower.

At that time, Pippin had a market cap of around $20 million, and not many people were paying attention. A market cap of $20 million also met my criteria for purchasing assets, as I prefer to buy in the $10-20-30 million range. I gradually bought 0.2% of the total supply, which I define as my maximum single position limit, spending about $40,000. It later dropped below $10 million, but I had already acquired my chips. The founder's background and technology do not change with price fluctuations. So after acquiring, I didn't pay much attention to it.

Later, Pippin announced a transformation to create a framework, shifting from a single AI Agent to an AI Framework, which led to a sharp increase in valuation. Even though the framework has not yet been developed, the recognition of the founder's technology and capabilities has propelled Pippin to a $300 million valuation. Creating a framework means it could potentially become a split project, and the market currently values frameworks at the highest in the AI agents sector, with the opportunity for a framework or ecosystem that can form a split project to exceed a $1 billion market cap.

Which AI Agents projects are you optimistic about? What do these projects do? Why are you optimistic?

defioasis: There are many projects I am optimistic about, like Pippin, which I still hold, but since its market cap is relatively high, I won't discuss it. I generally choose around the $20 million mark, but I actually have very few heavy investments.

Currently, there are mainly two directions: one is to mine gold from the Solana AI hackathon, which has now concluded, and there are quite a few award-winning projects that I am currently screening. The other is Virtuals moving to Solana and collaborating with Jupiter, which is expected to yield many interesting projects, as Virtuals has already proven its success on Base. Of course, I am still keeping an eye on this area.

Here, I mainly want to talk about some projects that emerged from the Solana hackathon. I will share one, but this is not investment advice. Recently, I have been observing a project called AgentiPy, which is also creating an open-source framework to connect AI agents to Solana's on-chain applications using Python. According to the roadmap, it may launch a self-narrative trading bot in Q1 and a launchpad in Q2; most importantly, it mentions that its token APY will participate as a flywheel. I studied the APY token economics, and it is well-designed; although it is also based on a fair launch similar to PumpFun, it has allocated 40% of its tokens and placed them all in Streamflow for a two-year linear unlock, indicating that the team has a certain level of commitment. The co-founders and CTO of AgentiPy have also been recognized by Solana's official Twitter. Emerging from the Solana hackathon, it at least serves as a form of endorsement. Of course, it is still quite early and has significant uncertainties. I will also pay more attention to the projects launched by Virtuals after it crosses over to Solana.

Overall, I believe AI is gradually moving towards the AI Application stage. Beyond frameworks, I will continue to focus on AI + applications, especially AI + DeFi, which combines the native narratives of AI and Crypto with DeFi assets and flywheels, potentially offering some good opportunities, but it is still in a very early stage, and I haven't seen any good targets yet. I am currently maintaining a watchful stance and haven't made any new asset purchases recently.

What is your view on the current AI Agents sector and market status? Do you think the popularity of AI Agents can last long, or do you believe this bubble has peaked?

defioasis: Recently, it has indeed been quite cool, but I don't think it will end just yet. AI still makes a lot of sense; AI outside the crypto space is rapidly iterating and developing, riding a wave of technological and capital momentum, which is a more important fundamental base. In fact, many AI targets are driven by external factors, whether in terms of narrative or talent. Shaw was once a marginal figure in web2, and now the ai16z he created has become a leader in Crypto AI. Influenced by Shaw, I believe more technical talents from traditional industries will come in to work on AI. Crypto AI itself is significantly lagging behind the outside world, so any wind or grass movement outside, as well as major updates, will transmit to Crypto, forming new narratives and sub-sectors.

From another perspective, within the crypto space, AI Agents are the only ones that have transitioned from general PumpFun to a scaled vertical sector. DeSci might count as half, and it has also cooled down recently. Apart from that, there are no other sectors that can transition from general to vertical, which actually indicates a strong demand for AI narratives. The current coolness of AI Agents is a cooling off from the previous overheating. Now, the proliferation of single Agents is forcing everyone to create frameworks, leading to growing fatigue towards such things. Therefore, AI + applications, especially AI + Crypto native narratives, if they can be established, I believe will reignite a wave of enthusiasm and new opportunities.

Regarding insights on investment and trading, what are some techniques for building and exiting positions?

defioasis: The above discussion has focused on the issue of investment targets, but I believe position management is even more important. Choosing targets is merely about technology, resources, and background. Although it is based on PumpFun for fair launches, the research approach is not much different from that of VC coins, involving technology, resources, background, team, endorsements, etc., as well as analysis of chip structures, mouse warehouse blue-chip addresses, and so on.

Now, let’s focus on the issue of position management. A decent asset can generally go through three stages: PvP, secondary stage, and listing, but most PvP projects disappear after that. I usually play in the secondary stage, focusing on assets in the $10-20-30 million market cap range, which I refer to as the on-chain sweet spot. I have found that some decent assets tend to consolidate and oscillate around this position after a surge and subsequent pullback. I generally only allocate assets within this range. I pay attention to those that have experienced one or more 70% pullbacks and are relatively stable within the $10-20-30 million market cap.

When I see a decent asset, I add it to my watchlist, categorizing them into S-level and A-level, among others. S-level assets are those that can form a pattern, such as being a parent coin ecosystem that can profit from continuously produced sub-coins. For example, as trading pair assets, sub-coins create a wealth effect, necessitating the parent coin for trading, thus generating continuous demand. This is similar to the SOL of PumpFun and the parent coin flywheel of Virtuals. Therefore, most S-level assets are framework types that have the potential to form ecosystems or pattern flywheels. The reason Pippin has been able to rise is that the market has high expectations for its framework development.

A-level assets are primarily judged based on narrative, background, technology, resources, and team (those that cannot form a pattern). The past background of the founder and founding team is very important, such as whether they have a background with the Solana Foundation, have many GitHub stars, have created products that have garnered attention, and whether the narrative has the potential to develop into a niche sector.

Generally speaking, I build positions in both A-level and S-level assets, with each position typically around $2,000 to $3,000. However, I am stricter about whether to increase my position or go heavy; I need to observe for a period to decide whether to add to my position, mainly looking at what the community and developers are doing daily.

I also have certain standards for the amount held in each coin. The maximum holding for a single coin is 0.2%, which means spending about $40,000 when purchasing around a $20 million market cap. Each time I decide whether to add to my position, I reflect on the target; if I determine that the target contradicts my initial purchase logic, I decisively abandon the idea of adding to my position, while temporarily keeping the remaining position pending; if it significantly drops below the market cap range, I also decisively abandon the idea of adding to my position, while temporarily keeping the remaining position. In principle, I consider selling only after achieving at least a 10x return, typically considering it in the $200-300 million market cap range.

The targets for building positions are often to ensure being on board, and I approach adding to positions cautiously, gradually increasing my holdings and setting a hard cap to avoid being overly blindly confident and getting stuck in a particular target. Of course, once I choose to add to a position, I will believe in my judgment of that target.

What I mentioned above are some strategies for the secondary stage. I feel that in the market, whether users, developers, or token factories seem to be more familiar with these secondary stage tactics, and it feels like the secondary stage has become relatively difficult to play recently. So I am also trying some lottery flow strategies to play a wave, which is essentially PvP. At this time, I am not limited to AI; I have prepared several SOL, buying in at 0.1 SOL each time, monitoring the market, and following some selected addresses, betting small to gain big. In the past couple of days, I have felt that in the current market, lottery flow strategies are easier to profit from. This might be a matter of probability; with so many projects, as long as the investment is small and diversified, and by monitoring some high-win-rate addresses as support, winning just once in the lottery flow can recover all previous losses. In the absence of a major market trend, this could be a good choice, and in some ways, it also prepares for potential major market movements. Maintaining a sense of profitability and the right mindset is very important.

免责声明：本文章仅代表作者个人观点，不代表本平台的立场和观点。本文章仅供信息分享，不构成对任何人的任何投资建议。用户与作者之间的任何争议，与本平台无关。如网页中刊载的文章或图片涉及侵权，请提供相关的权利证明和身份证明发送邮件到support@aicoin.com，本平台相关工作人员将会进行核查。