Base Ecology AI Deep Dive: Apart from Virtuals, What Other Hidden Gem Projects Are There?
Original Title: "Base Ecology AI Gold Rush: Beyond Virtuals, a Comprehensive Inventory of Those Hidden Gem Projects You May Have Overlooked"
Original Source: Deep Tide TechFlow

Foreword
Everyone praises AI Agent, and the Solana ecosystem is a treasure trove.
However, outside of Solana, Base is also one of the origins of this hype cycle. Do not forget that the market value of Virtuals is higher than that of ai16z.
Furthermore, one of the reasons why everyone likes the Base ecosystem is that its PvP intensity is not as high as Solana's, and you can also find many hidden gems there. However, due to scattered information, relatively low visibility of some projects, and information silos, seizing opportunities on Base is not that easy.
In addition to the tokens generated around Virtuals and Clanker in the past few months and the well-known AIXBT, what other hidden gem projects exist on Base now that you may have overlooked?
The editorial team has decided to start a major gold rush, attempting to find more AI Agent-related projects on Base to reveal the full landscape of opportunities in the entire ecosystem.
Based on different development directions and themes, we will list the recently outstanding projects here for your reference.
Rising Stars in Virtuals
In addition to AIXBT and the GAME framework, there are many potential projects in the Virtuals ecosystem with a market value of less than 100M that have performed well in recent months. Some of these noteworthy projects are as follows.
TAO CAT: AI Agent Built by Bittensor + Masa

$TAOCAT
CA: 0x7a5f5ccd46ebd7ac30615836d988ca3bd57412b3
Market Cap: 44M
Endorsed by well-known AI projects Masa and Bittensor, with excellent backgrounds and resources.
As a native AI agent in the Bittensor ecosystem, TAOCAT is directly built on top of Bittensor's subnets:
· SN42: Used for hyperintelligence real-time data.
· SN19: LLM/Reasoning, high-level language capabilities.
Functionally, TAOCAT is able to process and analyze real-time social data, combining AI technology for data interpretation and decision-making, providing market insights and analysis.
Some recent notable trends: Binance Alpha launch; TAOCAT also secured investment from DWF Labs, becoming a recipient of its $20 million AI agent fund.
Polytrader: AI-Powered Polymarket Prediction Assistant

$POLY
X: @polytraderAI
CA: 0x2676E4e0E2eB58D9Bdb5078358ff8A3a964CEdf5
Market Cap: 17M
Polytrader analyzes market sentiment and provides actionable insights to help make wiser, data-driven decisions on Polymarket.
You can have Polytrader analyze trending topics on social platforms, and then make a bet on the outcome of a particular event, essentially playing the role of a market predictor, akin to a vertically specialized AI agent that aligns with the previously popular Polymarket platform during the election period.

Specifically, to fully utilize this project, you need to hold a certain amount of POLY tokens, thus giving the token a defined utility.
During certain significant events, POLY may benefit not only from AI narratives but also from the events themselves.
Acolyte: An Oracle for AI Agents

X: @AcolytAI
CA: 0x79dacb99A8698052a9898E81Fdf883c29efb93cb
Market Cap: 37M
Acolyt provides reliable research and engineering data through infrastructure. It offers high-quality analysis and actionable insights for individuals and companies.
Currently, Acolyt is undergoing training to map and understand metrics that influence the artificial intelligence agent ecosystem, providing a comprehensive view of each agent and their role within it.
In the future, Acolyt will be a leading oracle, supplying high-quality data to artificial intelligence agents, traders, venture capital firms, index funds, AI accelerators, and educational companies. It will be capable of analyzing and generating probabilistic outcomes for any project, even before project launch.
Freya: AI Agent Integration with Gaming

$Freya
Twitter: @Freya_Starfall
CA: 0xF04D220b8136E2d3d4BE08081Dbb565c3c302FfD
Market Cap: 14M
Integrating AI agents into games to make in-game characters smarter and more imaginative has always been a hot topic.
Freya is a representative example of this concept. The token is associated with the popular Japanese-style game Starfall Chronicles (which utilizes Immutable's technology support) and allows leveraging AI capabilities in the game to enhance character interactions.
Therefore, Freya is not only a token but also an AI character in the game, positioned at the intersection of AI Agent and GameFi.
Recent Developments: The token has been listed on Binance Alpha; on January 6, the developer behind the game also participated in an AI16Z Japan meetup discussing Freya's freedom and future roadmap achieved through elizaTEE.

DeFAI
DeFAI refers to utilizing AI capabilities within DeFi to optimize various stages of DeFi transactions, before, during, and after, making it one of the key focal points discussed on social media recently.
For more insights on this track, please refer to the previous article: ""The "AI+DeFi" Trend Has Arrived, These DeFAI Sector Projects Are Worth Your Attention""
Kudai: DeFAI Agent Built by the GMX Community

$kudai
Website: @Kudai_IO
Contract Address: 0x288F4Eb27400fA220d14b864259Ad1B7f77C1594
Market Cap: 19M
Kudai is an AI agent born out of the GMX Blueberry Club community and built using the renowned framework @EmpyrealSDK, offering an experience that integrates community-driven spirit, DeFi, artificial intelligence, and innovative features.
It is important to note that Kudai has only released its token at the moment, with more capabilities yet to be formed; however, in the recent whitepaper release, it was mentioned that Kudai will later purchase and stake GMX to provide additional revenue streams, as well as invest in GMX's GM pool to further increase yield.
The project's envisioned goal is for Kudai to operate different strategies (leveraged trading, arbitrage, Farm negative interest rates, etc.) automatically on GMX V2, real-time share positions, profits, and losses.
It has a slight taste of GMX's AI image spokesperson, but more features are still on the way.
AI Agent Framework
REI: A New Framework for Efficient Blockchain and AI Collaboration
Website: @ReiNetwork0x
Contract Address: 0x6B2504A03ca4D43d0D73776F6aD46dAb2F2a4cFD
Market Cap: 104M
Rei Network is a core framework layer dedicated to maximizing the integration capabilities of AI and blockchain.
It has achieved a key goal through a three-layer architecture: enabling AI to be unrestricted by blockchain technology while being able to efficiently validate and store data on-chain at a low cost. What sets Rei apart is its ability to transform the probabilistic outputs of AI into deterministic, verifiable data structures.
Rei was the first project to adopt this approach, and with the first-mover advantage, they quickly established their leading position.

The technical details of framework-type projects are no longer redundant. The market generally believes that one benefit of REI is:
It separates blockchain and AI by design instead of forcibly combining them; it focuses on collaborative logic rather than fusion logic.
Currently, some AI Agents based on the Rei framework are also gradually emerging, but due to space constraints, they are not listed here. For a detailed analysis, see: "In-Depth Analysis of Rei Network and $REI: Another AI Project in the Base Ecosystem Not to Be Overlooked."
Investment/Incubation DAO
daos.world: A New Attempt at Decentralized Investment Funds

Following the popularity of the decentralized fund management platform Daos.fun launched by a16z, the concept of an "Investment DAO" began to be accepted --- a decentralized hedge fund managed by human or AI agents that raise funds, generate returns, and distribute profits back to DAO token holders.
On Base, a similar investment DAO is Daos.world.
Users can raise funds through this platform, easily launch and manage their own DAO hedge funds, and conduct trades through trustless smart contracts. Within each DAO, DAO managers raise ETH to start the fund. Managers can freely trade and invest ETH according to their preferences.
At the same time, each DAO also mints its own coin. Holding DAO tokens corresponds to "fund shares," which will later be used to share the investment returns of the fund.
Regarding the traded assets, the product homepage clearly states that these hedge funds are also related to Meme coins and AI.
Furthermore, the early products of daos.world are currently based on the Base chain, with plans to gradually expand to ETH, Hyperliquid, and others based on user demand.
Currently, four funds have been launched on daos.world, as outlined below.
Note: The DAO token in the daos.world ecosystem is currently experiencing a collective price drop. Please be aware of the risks and DYOR.

1. DR3AM DAO: AI-Assisted Investment Fund Targeting AI Opportunities
$FDREAM

X: @DR3AM_AI
CA: 0x0521AaA7C96E25afeE79FDd4f1Bb48F008aE4eac
Market Cap: 7M
DR3AM DAO is an AI-assisted investment fund targeting large, medium, and small-scale artificial intelligence opportunities. It is supported by the DREAM proprietary dataset and algorithm, combining human expertise with AI analysis to identify groundbreaking projects in the entire crypto AI space.
However, from DR3AM's current holdings, it appears to be mostly buying the tokens of other investment DAOs, such as WAI and TRUST (further details below).

2. Alchemist Accelerate: Holding Only BTC and ETH?
$ALCH (Not the one on Solana, please verify)

X: @ALCHDAO
CA: 0x2b0772BEa2757624287ffc7feB92D03aeAE6F12D
Market Cap: 1.2M
The DAO claims to invest in groundbreaking projects, create multilingual educational resources, and provide guidance and connections to enhance the global community's capabilities.
However, apart from holding the tokens of the aforementioned neighboring DAO FDREAM, ALCHDAO actually holds the most in Bitcoin and Ethereum, indicating a very conservative investment style...

3. AicroStrategy: On-chain Clone of MicroStrategy, Primarily Buying cbBTC
$AiSTR

X: @AicroStrategy
CA: 0x20ef84969f6d81Ff74AE4591c331858b20AD82CD
Market Cap: 4.2M
AicroStrategy is an AI hedge fund that will leverage cbBTC holdings to maximize Bitcoin exposure. The raised funds will be used to purchase cbBTC, which will be deployed to carefully selected DeFi protocols to maximize security and leverage.
The initial plan of the project is to deposit into Aave, borrow USDC, purchase more cbBTC, and then repeat the process. An AI algorithm will determine the optimal leverage ratio for executing the plan.
The DAO's holdings indeed reflect that it only buys cbBTC, but what it buys more of are the tokens of its sibling DAO $FDREAM.

4. Alameda Research V2: Playing the FTX Bankruptcy Joke, the Most Diversified AI Token Purchase
$AR

X: @AlamedaV2DAO
CA: 0x3e43cB385A6925986e7ea0f0dcdAEc06673d4e10
Market Cap: 1.5M
From the name and logo, it is evident that there is a strong sense of irony and humor, clearly paying homage to the FTX bankruptcy event and the Alameda Research associated with SBF and his girlfriend.
The DAO's introduction is also very interesting: "V2 is a leading trading firm operated from SBF's cell... We use internally developed technology and deep expertise of the team in crypto to trade thousands of digital asset products."
Looking at the fund's holdings, it is indeed the most diversified and spread out among several DAOs, basically including the popular AI-related tokens on Base.

wai combinator: On-chain Version of Y Combinator Homophonic Pun, Base's AI Project Incubator
$WAI

X: @wai_combinator
CA: 0x6112b8714221bBd96AE0A0032A683E38B475d06C
Market Cap: 17M
The WAI Combinator is actually a project built on the Virtuals protocol, but more inclined towards investment incubation.
Its core positioning is as an experimental "Agent + Human" investment DAO organization, aiming to create value for Token holders by making investment decisions through a combination of AI agents and human expertise.
From the name, WAI is clearly paying tribute to the well-known tech company incubator Y Combinator, and its business is similar. However, the projects incubated are all within the on-chain Base ecosystem, especially early-stage projects within Virtuals still in the Bonding Curve phase.
Its asset under management has grown from 50k to over 700k USD in less than two weeks, with the portfolio value showing significant growth (currently at 500k USD). At the same time, the project is continuously deploying new investments through the "Velocity" plan, so staying tuned for projects that it favors is advisable.
WAI has also disclosed its asset management address (click here), and interested players can track it.

Similarly, holding WAI tokens allows you, like other investment DAOs, to share in the organization's investment fund's returns. However, compared to the aforementioned investment DAO, the WAI Combinator has more business opportunities, such as:
Providing AI technical support and integration to help projects optimize decision-making processes; assisting projects in accessing the Virtuals ecosystem to promote collaboration and resource sharing among projects, and so forth.
So it is more like a strategic investment + incubation for listed projects role, rather than just making investments.
Note: A similar investment DAO project to WAI is Vader AI, which is often compared by Key Opinion Leaders (KOLs):
$VADER
CA: 0x731814e491571A2e9eE3c5b1F7f3b962eE8f4870
trustmebrosfun: Incubated by WAI Combinator, about to launch a new coin $DATDAO
$TRUST

Telegram: @trustmebrosfun
CA: 0xC841b4eaD3F70bE99472FFdB88E5c3C7aF6A481a
Market Cap: 12M
The above-mentioned WAI Combinator has successfully incubated a project called trustmebrosfun, an AI project running on the Base chain. The name comes from a common meme in the cryptocurrency community "Trust me bro," which ironically emphasizes the trust issue in the crypto market.
The existing token of the project is $TRUST, which is both a meme and corresponds to the social media AI Agent of trust.
However, the key point is that its founder, @Degen__Ape__, recently released the DATDAO whitepaper based on trustmebrosfun. DATDAO is defined as an innovative decentralized autonomous organization that will focus on investing in other tokens and PVP airdrop mining on Hyperliquid.
DATDAO will soon launch its own token $DATDAO, and to qualify for the whitelist to obtain this token, a minimum of 100K TRUST tokens is required in the public round, which will correspond to an allocation of $DATDAO worth 1 ETH, with a personal cap of 5 ETH.
For more rules, please refer to the founder's original post.
Setting aside questions about DATDAO's investment level, whether it generates returns, and the market need for another investment DAO, from a tokenomics perspective, participation in DATDAO requires holding TRUST. This may create some short-term demand for TRUST.
sekoia_virtuals: On-Chain Version of Sequoia Capital
$ SEKOIA
X: @sekoia_virtuals

CA: 0x1185cB5122Edad199BdBC0cbd7a0457E448f23c7
Market Cap: 60M
This project also exists on Virtuals, but the author classified it under an investment DAO or on-chain fund.
SEKOIA aims to build the best-performing on-chain venture capital agency. The project's X uses a semi-automatic, semi-manual AI posting approach and claims to aim to outperform traditional companies for better outcomes.
From its name, it's clear that it pays tribute to Sequoia, i.e., Sequoia Capital.
During the recent crypto market downturn, the project's token price remained relatively stable compared to other mentioned investment DAOs. This stability may stem from the fund's investment in another token named $VOLTX, an AI Agent token, which has shown excellent performance and gained market confidence.
Based on information displayed on its website, the on-chain Sequoia's investment level is quite good, achieving profits 15 times its investment total.

AI Agent Launchpad
CreatorBid - A New Launchpad Integrating Bittensor and Autonolas Capabilities

Capital that has benefited from the rise of Virtuals will inevitably have a spill-over effect, looking for more distinctive small-scale ecosystems within the same Base.
Within this context, CreatorBid is certainly an essential part. Its most significant feature lies in integration with the Bittensor subnet and Olas, bringing significant utility to its ecosystem, especially through Olas' upcoming Mech Marketplace.
In simple terms, Olas' Mech Marketplace is like an "Agent Marketplace," where agents can autonomously acquire new skills, tools, and workflows—expanding their capabilities in real time. This technology enables CreatorBid to meet the growing demand for agents focusing on predictive workflows, transactions, payment automation, and more.

Therefore, you can think of CreatorBid as a creation platform that combines blockchain and AI technologies to provide content creators with AI model development and deployment support, as well as offer unhosted platform services.
Technically, it collaborates with io.net and Aethir's GPU network for AI model scaling; simultaneously leveraging Bittensor's subnet capabilities and Olas' Agent Marketplace prowess to jointly build its platform.
As the platform's token $BID has not yet undergone a TGE, but has already built 70+ Agents, we can focus on reviewing the following 3 projects within the ecosystem and their associated tokens:
1. draiftking: Using machine learning to analyze soccer matches and guide sports betting
$DKING

X: @draiftking
Market Cap: 32M
CA: 0x57eDc3F1fd42c0D48230e964b1C5184B9c89B2ed
This project, developed by @webuildscore, uses machine learning to analyze vast amounts of data such as player positions, match-ups, performance status, etc., to quickly identify inefficiencies in the sports betting market and place profitable bets.

Behind the project, the CEO is a member of @crunchDAO, a top-tier machine learning community, and the team members have extensive experience in sports analysis and the gambling market.
It is worth highlighting CrunchDAO, a mature elite machine learning community overseas (over 7,000 data scientists - over 700 PhDs), which develops Alpha insights through its collective intelligence network.
At the same time, this AI agent relies on the TAO Network and, based on the ScoreVision subnet (SN44), can significantly reduce computing costs, meaning ScoreVision and DraftKings have a certain technological moat in competition.
For more detailed content, you can refer to the detailed analysis post by renowned Alpha blogger @SmallCapScience.
2. Eolas: Tool Enhancing AI Agents on the CreatorBid Platform
$EOLAS

Twitter: @Eolas_AI
Market Cap: 7.6M
Contract Address: 0xF878e27aFB649744EEC3c5c0d03bc9335703CFE3
Eolas is a tool system focused on enhancing AI agents on the CreatorBid platform and is the first agent on the platform to utilize Autonolas capabilities.
In simple terms, Eolas's main advantage is that it makes advanced features that were previously only available to a few elite AI agents (such as @aixbt_agent) accessible to everyone, thus enhancing the overall capability of the CreatorBid ecosystem.
Its main function is to provide advanced tools and features for AI agents on CreatorBid. Eolas uses Autonolas's Olas Mechs technology to develop and distribute these tools.
Among them, Olas Mech is a core technology in the Autonolas network, which is essentially a blockchain-based AI service marketplace. This technology allows AI agents (referred to as Mechs) to provide AI services to other agents or applications via the blockchain.
Returning to Eolas, the core idea of this system is to create and share AI tools in a decentralized manner. Eolas claims that this approach can elevate the overall capability of the CreatorBid ecosystem. It also includes an economic model to sustain the operation and development of the system through tool usage fees and development rewards.
3. AION 5100: Prediction Agent on the Bittensor Subnet
$AION

X: @aion5100
Market Cap: 40M
CA: 0xfc48314ad4ad5bd36a84e8307b86a68a01d95d9c
AION 5100 is an AI Agent project developed by CreatorBid running on the Bittensor network, with its core functionality being to provide self-improving prediction services on the Bittensor prediction subnet (Subnet 6).
The project uses the TAO token as its underlying support. While this AI Agent always attracts attention through a mystical marketing approach such as speaking as the "AI from the future," its essence lies in being an application project focused on AI prediction capabilities, particularly in predicting financial and market trends.

It's worth mentioning that the project's official website is quite interesting, featuring a sleek computer where clicking different buttons prompts messages about various functionalities being prepared. However, the website currently lacks substantial content and functionality, leaning more towards marketing and intentions.
Regarding what AION5000 specifically can predict, there is no clear indication. The Agent account on X is currently engaging in abstract and non-specific self-talk about predictions, and its future performance will need further observation.
Airdrop and Beta Opportunity: Native Coin $BID, $TAO, and $OLAS
Aside from holding different Agent tokens, the platform's native coin $BID has yet to undergo a TGE. Therefore, a potential opportunity is to lock up the Agent tokens purchased on Creator.Bid to receive the platform's daily distributed points, which can be later exchanged for $BID at a certain ratio.

Beta-related opportunities:
TAO: Some projects benefit from Bittensor's subnet, which can be seen as a positive development to some extent.
OLAS: As Creatorbid has integrated Autonolas' Mech capability, OLAS itself may also benefit from this platform's hype.
Loomlay -- Building an AI Agent in a No-Code Environment
$LAY

Telegram: @loomlayai
Contract Address: 0xb89d354ad1b0d95a48b3de4607f75a8cd710c1ba
Market Cap: 130M
Loomlay is also an innovative platform that combines AI agent collaboration with Web3 technology.
The platform aims to simplify the process of creating and deploying AI agents, allowing more users to participate in AI application development through a no-code environment.

The Loomlay agent consists of three core components—Model, Plugins, and Workflows—which together create a fully functional AI system.
In terms of agent building, Loomlay employs an intuitive visual interface where users can define the capabilities and objectives of the AI agent through simple configurations. The system integrates ERC4337 wallet functionality to provide infrastructure support for agent's autonomous operations.
The platform's plugin ecosystem greatly expands the agent's application scope. By integrating different types of plugins, AI agents can perform a variety of tasks, from market analysis to content creation. This modular design allows agents to flexibly adjust their functionality based on actual needs.
On the collaboration front, Loomlay has established a comprehensive AI agent collaboration network. The platform supports combining agents with different expertise, such as pairing market analysis agents with trading agents or connecting research-oriented agents with content creation agents to achieve collaborative processing of complex tasks.
The platform's native token $LAY incorporates a carefully designed deflationary mechanism.
In token transactions on Uniswap V3, 1% of the daily sell order amount is used for token burning. Additionally, 20% of the $LAY payment volume in market transactions is permanently removed from circulation.

It is worth mentioning that Loomlay is relatively new. The platform was launched on December 31st, and within a week, it had already created over 500 agents, with more than 70 of them already having tokens.
However, due to its early stage, the Agent tokens have not seen significant gains yet, and we will continue to observe further.
Exploring More Directions and Filling the Gaps
Our editors' resources are also very limited, and the above cannot cover all AI Agent projects in the entire Base ecosystem.
For more information, you can refer to a chart created by renowned blogger @sandraaleow to group and integrate AI projects in different directions to understand where the AI projects in the current Base ecosystem are heading.

Additionally, some projects have not issued tokens yet, or they have migrated old businesses to Base. You can also refer to the table below:

Chart: Original post from @davidtsocy, compiled and tabulated by Deep Tide TechFlow
Lastly, it is important to note that the risk in the AI Agent trend should not be ignored, as it is questionable whether many projects actually have AI support. All projects listed in this article do not constitute any investment advice, and for more information, please do your research (DYOR).
The above text is just a start, and we hope everyone can find more of their own Alpha.
猜你喜歡
穩定幣驅動全球B2B支付革新,如何打破工作流程瓶頸釋放兆市場潛力?
這些新創公司正在無需資料中心的情況下建立先進AI模型
CEX與Wallet之後,OKX入局支付
RWA永續產品危機:為什麼GLP模式註定撐不住RWA永續?
科學平權運動:DeSci的萬億美元知識經濟重建革命
Sentient深度研報:獲8,500萬美元融資,建置去中心化AGI新範式
加密項目為何札堆發卡?一場Web3與現實世界的通行權之爭
專訪Virtuals聯創empty:AI 創業不需要大量資金,Crypto是答案之一
今年 2 月,Base 生態中的 AI 協議 Virtuals 宣布跨鏈至 Solana,然而加密市場隨後進入流動性緊縮期,AI Agent 板塊從人聲鼎沸轉為低迷,Virtuals 生態也陷入一段蟄伏期。
三月初,BlockBeats 對 Virtuals 共同創辦人 empty 進行了一次專訪。彼時,團隊尚未推出如今被廣泛討論的 Genesis Launch 機制,但已在內部持續探索如何透過機制設計激活舊資產、提高用戶參與度,並重構代幣發行與融資路徑。那是一個市場尚未復甦、生態尚處冷啟動階段的時間點,Virtuals 團隊卻沒有停下腳步,而是在努力尋找新的產品方向和敘事突破口。
兩個月過去,AI Agent 板塊重新升溫,Virtuals 代幣反彈超 150%,Genesis 機製成為帶動生態回暖的重要觸發器。從積分獲取規則的動態調整,到專案參與熱度的持續上升,再到「新代幣帶老代幣」的機制閉環,Virtuals 逐漸走出寒冬,並再次站上討論焦點。
值得注意的是,Virtuals 的 Genesis 機制與近期 Binance 推出的 Alpha 積分系統有一些相似之處,評估用戶在 Alpha 和幣安錢包生態系統內的參與度,決定用戶 Alpha 代幣空投的資格。用戶可透過持倉、交易等方式獲得積分,積分越高,參與新項目的機會越大。透過積分系統篩選使用者、分配資源,專案方能夠更有效地激勵社群參與,提升專案的公平性和透明度。 Virtuals 和 Binance 的探索,或許預示著加密融資的新趨勢正在形成。
回看這次對話,empty 在專訪中所展現出的思路與判斷,正在一步步顯現其前瞻性,這不僅是一場圍繞打新機制的訪談,更是一次關於“資產驅動型 AI 協議”的路徑構建與底層邏輯的深度討論。
BlockBeats:可以簡單分享一下最近團隊主要在忙些什麼?
empty:目前我們的工作重點主要有兩個部分。第一部分,我們希望將 Virtuals 打造成一個類似「華爾街」的代理人(Agent)服務平台。設想一下,如果你是專注於 Agent 或 Agent 團隊建立的創業者,從融資、發幣到流動性退出,整個流程都需要係統性的支援。我們希望為真正專注於 Agent 和 AI 研發的團隊,提供這一整套服務體系,讓他們可以把精力集中在底層能力的開發上,而不用為其他環節分心。這一塊的工作其實也包括了與散戶買賣相關的內容,後面可以再詳細展開。
第二部分,我們正在深入推進 AI 相關的佈局。我們的願景是建立一個 AI 社會,希望每個 Agent 都能聚焦自身優勢,同時透過彼此之間的協作,實現更大的價值。因此,最近我們發布了一個新的標準——ACP(Agent Communication Protocol),目的是讓不同的 Agent 能夠相互互動、協作,共同推動各自的業務目標。這是目前我們主要在推進的兩大方向。
BlockBeats:可以再展開說說嗎?
empty:在我看來,其實我們面對的客戶群可以分為三類:第一類是專注於開發 Agent 的團隊;第二類是投資者,包括散戶、基金等各種投資機構;第三類則是 C 端用戶,也就是最終使用 Agent 產品的個人用戶。
不過,我們主要的精力其實是放在前兩大類──也就是團隊和投資人。對於 C 端用戶這一塊,我們並不打算直接介入,而是希望各個 Agent 團隊能夠自己解決 C 端市場的拓展問題。
此外,我們也認為,Agent 與 Agent 之間的交互作用應該成為一個核心模式。簡單來說,就是未來的服務更多應該是由一個 Agent 銷售或提供給另一個 Agent,而不是單純賣給人類使用者。因此,在團隊的 BD 工作中,我們也積極幫助現有的 AI 團隊尋找這樣的客戶和合作機會。
BlockBeats:大概有一些什麼具體案例呢?
empty:「華爾街」說白了就是圍繞資本運作體系的建設,假設你是一個技術團隊,想要融資,傳統路徑是去找 VC 募資,拿到資金後開始發展。如果專案做得不錯,接下來可能會考慮進入二級市場,例如在紐約證券交易所上市,或是在 Binance 這樣的交易所上幣,實現流動性退出。
我們希望把這一整套流程打通-從早期融資,到專案開發過程中對資金的靈活使用需求,再到最終二級市場的流動性退出,全部覆蓋和完善,這是我們希望補齊的一條完整鏈條。
而這一部分的工作和 ACP(Agent Communication Protocol)是不同的,ACP 更多是關於 Agent 與 Agent 之間交互標準的製定,不直接涉及資本運作系統。
BlockBeats:它和現在 Virtuals 的這個 Launchpad 有什麼差別呢?資金也是從 C 端來是嗎?
empty:其實現在你在 Virtuals 上發幣,如果沒有真正融到資金,那就只是發了一個幣而已,實際是融不到錢的。我們目前能提供的服務,是透過設定買賣時的交易稅機制,從中提取一部分稅收回饋給創業者,希望這部分能成為他們的現金流來源。
不過,問題其實還分成兩塊。第一是如何真正幫助團隊完成融資,這個問題目前我們還沒有徹底解決。第二是關於目前專案發行模式本身存在的結構性問題。簡單來說,現在的版本有點像過去 Pumpfun 那種模式——也就是當專案剛上線時,部分籌碼就被外賣給了外部投資人。但現實是,目前整個市場上存在著太多機構集團和「狙擊手」。
當一個真正優秀的專案一發幣,還沒真正觸達普通散戶,就已經被機構在極高估值時搶購了。等到散戶能夠接觸到時,往往價格已經偏高,專案品質也可能變差,整個價值發行體係被扭曲。
針對這個問題,我們希望探索一種新的發幣和融資模式,目的是讓專案方的籌碼既不是死死握在自己手裡,也不是優先流向英文圈的大機構,而是能夠真正留給那些相信專案、願意長期支持專案的普通投資者手中。我們正在思考該如何設計這樣一個新的發行機制,來解決這個根本問題。
BlockBeats:新模式的具體想法會是什麼樣子呢?
empty:關於資金這一塊,其實我們目前還沒有完全想透。現階段來看,最直接的方式還是去找 VC 融資,或是採取公開預售等形式進行資金募集。不過說實話,我個人對傳統的公開預售模式並不是特別認同。
在「公平發售」這件事上,我們正在嘗試換一個角度來思考-希望能從「reputation」出發,重新設計機制。
具體來說,就是如果你對整個 Virtuals 生態有貢獻,例如早期參與、提供支持或建設,那麼你就可以在後續購買優質代幣時享有更高的優先權。透過這種方式,我們希望把資源更多留給真正支持生態發展的用戶,而不是由短期套利的人主導。
BlockBeats:您會不會考慮採用類似之前 Fjord Foundry 推出的 LBP 模式,或者像 Daos.fun 那種採用白名單機制的模式。這些模式在某種程度上,和您剛才提到的「對生態有貢獻的人享有優先權」的想法是有些相似的。不過,這類做法後來也引發了一些爭議,例如白名單內部操作、分配不公等問題。 Virtuals 在設計時會考慮借鏡這些模式的優點,或有針對性地規避類似的問題嗎?
empty:我認為白名單機制最大的問題在於,白名單的選擇權掌握在專案方手中。這和「老鼠倉」行為非常相似。專案方可以選擇將白名單名額分配給自己人或身邊的朋友,導致最終的籌碼仍然掌握在少數人手中。
我們希望做的,依然是類似白名單的機制,但不同的是,白名單的獲取權應基於一個公開透明的規則體系,而不是由項目方單方面決定。只有這樣,才能真正做到公平分配,避免內幕操作的問題。
我認為在今天這個 AI 時代,很多時候創業並不需要大量資金。我常跟團隊強調,你們應該優先考慮自力更生,例如透過組成社區,而不是一開始就想著去融資。因為一旦融資,實際上就等於背負了負債。
我們更希望從 Training Fee的角度去看待早期發展路徑。也就是說,專案可以選擇直接發幣,透過交易稅所帶來的現金流,支持日常營運。這樣一來,專案可以在公開建設的過程中獲得初步資金,而不是依賴外部投資。如果專案做大了,自然也會有機會透過二級市場流動性退出。
當然最理想的情況是,專案本身能夠有穩定的現金流來源,這樣甚至連自己的幣都無需拋售,這才是真正健康可持續的狀態。
我自己也常在和團隊交流時分享這種思路,很有意思的是,那些真正抱著「搞快錢」心態的項目,一聽到這種機制就失去了興趣。他們會覺得,在這種模式下,既無法操作老鼠倉,也很難短期套利,於是很快就選擇離開。
但從我們的角度來看,這其實反而是個很好的篩選機制。透過這種方式,理念不同的專案自然會被過濾出去,最後留下的,都是那些願意真正建立、和我們價值觀契合的團隊,一起把事情做起來。
BlockBeats:這個理念可以發展出一些能夠創造收益的 AI agent。
empty:我覺得這是很有必要的。坦白說,放眼今天的市場,真正擁有穩定現金流的產品幾乎鳳毛麟角,但我認為這並不意味著我們應該停止嘗試。事實上,我們每天在對接的團隊中,有至少一半以上的人依然懷抱著長遠的願景。很多時候,他們甚至已經提前向我們提供了 VC 階段的資金支持,或表達了強烈的合作意願。
其實對他們來說想要去收穫一個很好的社區,因為社區可以給他們的產品做更好的回饋,這才是他們真正的目的。這樣聽起來有一點匪夷所思,但其實真的有很多這樣的團隊,而那種團隊的是我們真的想扶持的團隊。
BlockBeats:您剛才提到的這套「AI 華爾街」的產品體系-從融資、發行到退出,建構的是一整套完整的流程。這套機制是否更多是為了激勵那些有意願發幣的團隊?還是說,它在設計上也考慮瞭如何更好地支持那些希望透過產品本身的現金流來發展的團隊?這兩類團隊在您這套體系中會不會被區別對待,或者說有什麼機制設計能讓不同路徑的創業者都能被合理支持?
empty:是的,我們 BD 的核心職責其實就是去鼓勵團隊發幣。說得直接一點,就是引導他們思考發幣的可能性和意義。所以團隊最常問的問題就是:「為什麼要發幣?」這時我們需要採取不同的方式和角度,去幫助他們理解背後的價值邏輯。當然如果最終判斷不適合,我們也不會強迫他們推進。
不過我們觀察到一個非常明顯的趨勢,傳統的融資路徑已經越來越難走通了。過去那種融資做大,發幣上所的模式已經逐漸失效。面對這樣的現實,很多團隊都陷入了尷尬的境地。而我們希望能從鏈上和加密的視角,提供一套不同的解決方案,讓他們找到新的發展路徑。
BlockBeats:明白,我剛才其實想表達的是,您剛剛也提到,傳統的 AI 模式在很大程度上仍然依賴「燒錢」競爭。但在 DeepSeek 出現之後,市場上一些資金體積較小的團隊或投資人開始重新燃起了信心,躍躍欲試地進入這個領域。您怎麼看待這種現象?這會不會對目前正在做 AI 基礎研發,或是 AI 應用層開發的團隊產生一定的影響?
empty:對,我覺得先不談 DeepSeek,從傳統角度來看,其實到目前為止,AI 領域真正賺錢的只有英偉達,其他幾乎所有玩家都還沒有實現盈利。所以其實沒有人真正享受了這個商業模式的成果,大家也仍在探索如何面對 C 端打造真正有產出的應用。
沒有哪個領域像幣圈一樣能如此快速獲得社群回饋。你一發幣,用戶就會主動去讀白皮書的每一個字,試試你產品的每個功能。
當然,這套機制並不適合所有人。例如有些 Agent 產品偏 Web2,對於幣圈用戶而言,可能感知不到其價值。因此,我也會鼓勵做 Agent 的團隊在 Virtuals 生態中認真思考,如何真正將 Crypto 作為自身產品的差異化要素加以運用與設計。
BlockBeats:這點我特別認同,在 Crypto 這個領域 AI 的迭代速度確實非常快,但這群用戶給予的回饋,真的是代表真實的市場需求嗎?或者說這些回饋是否真的符合更大眾化、更具規模性的需求?
empty:我覺得很多時候產品本身不應該是強行推廣給不適合的使用者群體。例如 AIXBT 最成功的一點就在於,它的用戶本身就是那群炒作他人內容的人,所以他們的使用行為是非常自然的,並不覺得是在被迫使用一個無聊的產品。 mass adoption 這個概念已經講了很多年,大家可能早就該放棄這個執念了。我們不如就認了,把東西賣給幣圈的人就好了。
BlockBeats:AI Agent 與 AI Agent 所對應的代幣之間,究竟應該是什麼樣的動態關係?
empty:對,我覺得這裡可以分成兩個核心點。首先其實不是在投資某個具體的 AI Agent,而是在投資背後經營這個 Agent 的團隊。你應該把它理解為一種更接近創投的思路:你投的是這個人,而不是他目前正在做的產品。因為產品本身是可以快速變化的,可能一個月後團隊會發現方向不對,立即調整。所以,這裡的「幣」本質上代表的是對團隊的信任,而不是某個特定 Agent 本身。
第二則是期望一旦某個 Agent 產品做出來後,未來它能真正產生現金流,或者有實際的使用場景(utility),從而讓對應的代幣具備賦能效應。
BlockBeats:您覺得有哪些賦能方式是目前還沒看到的,但未來可能出現、值得期待的?
empty:其實主要有兩塊,第一是比較常見的那種你要使用我的產品,就必須付費,或者使用代幣支付,從而間接實現對代幣的「軟銷毀」或消耗。
但我覺得更有趣的賦能方式,其實是在獲客成本的角度思考。也就是說,你希望你的用戶同時也是你的投資者,這樣他們就有動機去主動幫你推廣、吸引更多用戶。
BlockBeats:那基於這些觀點,您怎麼看 ai16z,在專案設計和代幣機制方面,似乎整體表現並不太樂觀?
empty:從一個很純粹的投資角度來看,撇開我們與他們之間的關係,其實很簡單。他們現在做的事情,對代幣本身沒有任何賦能。從開源的角度來看,一個開源模型本身是無法直接賦能代幣的。
但它仍然有價值的原因在於,它像一個期權(call option),也就是說,如果有一天他們突然決定要做一些事情,比如推出一個 launchpad,那麼那些提前知道、提前參與的人,可能會因此受益。
開發者未來確實有可能會使用他們的 Launchpad,只有在那一刻,代幣才會真正產生賦能。這是目前最大的一個問號——如果這個模式真的跑得通,我認為確實會非常強大,因為他們的確觸達了大量開發者。
但我個人還是有很多疑問。例如即使我是使用 Eliza 的開發者,也不代表我一定會選擇在他們的 Launchpad 上發幣。我會貨比三家,會比較。而且,做一個 Launchpad 和做一個開源框架,所需的產品能力和社群運作能力是完全不同的,這是另一個重要的不確定性。
BlockBeats:這種不同是體現在什麼地方呢?
empty:在 Virtuals 上我們幾乎每天都在處理客服相關的問題,只要有任何一個團隊在我們平台上發生 rug,即使與我們沒有直接關係,用戶也會第一時間來找我們投訴。
這時我們就必須出面安撫用戶,並思考如何降低 rug 的整體風險。一旦有團隊因為自己的代幣設計錯誤或技術失誤而被駭客攻擊、資產被盜,我們往往需要自掏腰包,確保他們的社群至少能拿回一點資金,以便專案能夠重新開始。這些項目方可能在技術上很強,但未必擅長代幣發行,結果因操作失誤被攻擊導致資產損失。只要涉及「被欺騙」相關的問題,對我們來說就已經是非常麻煩的事了,做這些工作跟做交易所的客服沒有太大差別。
另一方面,做 BD 也非常困難。優秀的團隊手上有很多選擇,他們可以選擇在 Pumpfun 或交易所上發幣,為什麼他們要來找我們,那這背後必須要有一整套支援體系,包括融資支援、技術協助、市場推廣等,每個環節都不能出問題。
BlockBeats:那我們就繼續沿著這個話題聊聊 Virtuals 目前的 Launchpad 業務。有一些社群成員在 Twitter 上統計了 Virtuals Launchpad 的整體獲利狀況,確實目前看起來獲利的項目比較少。接下來 Launchpad 還會是 Virtuals 的主要業務區嗎?還是說,未來的重心會逐漸轉向您剛才提到的「AI 華爾街」這條路徑?
empty:其實這兩塊本質上是一件事,是一整套體系的一部分,所以我們必須繼續推進。市場的波動是很正常的,我們始終要堅持的一點是:非常清楚地認識到我們的核心客戶是誰。我一直強調我們的客戶只有兩類——團隊。所以市場行情的好壞對我們來說並不是最重要的,關鍵是在每一個關鍵節點上,對於一個團隊來說,發幣的最佳選擇是否依然是我們 Virtuals。
BlockBeats:您會不會擔心「Crypto + AI」或「Crypto AI Agent」這一類敘事已經過去了?如果未來還有一輪多頭市場,您是否認為市場炒作的焦點可能已經不再是這些方向了?
empty:有可能啊,我覺得 it is what it is,這確實是有可能發生的,但這也屬於我們無法控制的範圍。不過如果你問我,在所有可能的趨勢中,哪個賽道更有機會長期保持領先,我仍然認為是 AI。從一個打德撲的角度來看,它仍然是最優選擇。
而且我們團隊的技術架構和底層能力其實早已搭建完成了,現在只是順勢而為而已。更重要的是,我們本身真的熱愛這件事,帶著好奇心去做這件事。每天早上醒來就有驅動力去研究最新的技術,這種狀態本身就挺讓人滿足的,對吧?
很多時候,大家不應該只看產品本身。實際上很多優秀的團隊,他們的基因決定了他們有在規則中勝出的能力——他們可能過去在做派盤交易時,每筆規模就是上百萬的操作,而這些團隊的 CEO,一年的薪資可能就有 100 萬美金。如果他們願意出來單幹項目,從天使投資或 VC 的視角來看,這本質上是用一個很划算的價格買到一個高品質的團隊。
更何況這些資產是 liquid 的,不是鎖倉狀態。如果你當下不急著用錢,完全可以在早期階段買進一些優秀團隊的代幣,靜靜等待他們去創造一些奇蹟,基本上就是這樣一個邏輯。