From Hype to Practical: Value Transformation of Web3 AI Agents

Original title:Post-AI Agent Bubble: Where's the Real Value in Web3 AI?

Original author: 0x Jeff

Original translation: Ethan, Odaily Planet Daily

Brief Overview:

The total market value of AI agents (Agents) grew from 0 to over $20 billion in a few months, and then quickly collapsed. But this field is gradually maturing. Infrastructure, decentralized AI, and real practicality are gradually taking over. How will the next wave of development shape the future and why is it worth paying attention to?

In the fourth quarter of last year, we saw a fastest growing field, “AI Agent,” which grew from zero to over $20 billion in just a few months – from some fun, charismatic, entertaining “agents” to financial agents that promise to change the world and get rich overnight through transactions and investments. And…not just those agents that make you rich, investment DAOs have also emerged… Human (or agents) DAOs (3, 3) invest in other agents.

From Hype to Infrastructure: The Evolution of Web3 AI Agents

We all understand that in an emerging field (and in the context of Web2 AI, Trump’s election and support for new catalysts like cryptocurrencies and AI), people don’t care about fundamentals. Any demonstration that can make noise, looks hype, cool can almost quickly reach a market value of over $100 million.

@virtuals_io has become an ecosystem that occupys the market, captures market attention, tells good stories, and creates the best narrative. This has attracted a large number of creators to publish content on Virtuals and launch projects, while attracting the attention of retail investors to capitalize hype.

@elizaOS then emerged, taking a different approach -open source AI, allowing any developer to use a "shovel" to dig out gold. A huge spread effect around this concept has been formed, adopting rapid growth, and the number of stars and branches on GitHub has also increased dramatically (these numbers are still growing).

Virtuals has grown to over $5 billion, and Eliza accounts for about half of the market cap at its highest point, with some other interesting agents reaching 8-9-digit valuations, such as AIXBT reaching $1 billion. Of course, the situation is very different now, with newly released, well-performing agents having an average valuation of between $3 million and $10 million. Old, well-performing agents have averaged between $10 million and $50 million. The valuation ceiling was compressed, and the overall market value also narrowed from US$20 billion to US$4 billion to US$6 billion.

Infrastructure acceleration and rapid advancement of Web2

The market is now beginning to focus on "pure fundamentals", and people are more concerned about infrastructure, decentralized AI, especially the AI ​​models in Web2 continue to accelerate at an astonishing speed - Meta's Llama, OpenAI's GPT, Grok, DeepSeek, Alibaba Qwen and others will release new and more optimized models every month. ChatGPT's image generation model quickly triggered a "giplier" trend after its release.

Above all this, the consumer layer of Web2 has developed much faster than before due to the improvement of AI model capabilities - things that were impossible to achieve before have now become possible. Tools such as Lovable, Bolt, Cursor and Windsurf enable developers to launch more products faster. Agent workflows and AI agents are everywhere. The entry barrier is lowered, and the switching cost of users is almost zero - if you hate an application, you can easily find more competitive services or products, with a better interface and user experience.

Data ownership awakening: The call for decentralized AI

Meanwhile, many people began to think: "Since so many proxy applications use centralized technology, who owns my data? Where will my data go? If I talked about something private with AI, will it keep it private? Or will it leak out?" This question is particularly important, especially since the recent update of OpenAI mentioned that ChatGPT is now able to cite your past chat history to provide a more personalized response.

Alas… it all sounds cool and is likely to trigger a wave of personalized AI agents like co-pilots, personal secretary, therapist, partner, etc. But you can imagine what the consequences will be when someone has or has mastered your data.

Decentralized AI (DeAI): The Power to Lead the Future

I made some predictions last year, one of which is that decentralized AI will make its mark in the second quarter of 2025, with infrastructure increasing confidentiality, transparency, verifiability and data ownership and gaining more adoption and attention as a result.

This trend can be divided into three main parts, many of which intersect or intertwined:

• Web2 AI venture capital trends (YC company launches vertical agents, a16z makes layouts for future consumer trends, and Perplexity launches AI fund)

• Web3 AI venture capital trends (DeAI infrastructure investment, distributed training, inference networks, etc.)

• Web3 AI retail trends (AI agent ecosystem, consumer agent, AI consumer applications)

Web2 and Web3 AI: The Collision of Two Worlds

For Web2 , since the total address market (TAM) is much larger than Web3, many businesses are looking to transform and optimize their businesses through AI and improve workflows to generate more leads, more conversions, higher sales, retain more customers, reduce administrative costs, and operate in a more efficient way. Therefore, many businesses seek solutions that can solve their specific pain points.

This need for optimization has attracted many young entrepreneurs to find better ways to bring AI agents into their workflow. Compared to traditional SaaS, AI agents provide solutions that can save a lot of money or generate more leads. This allows agency startups to charge higher subscription fees for their use (which is why we see many startups reach 7-8-figure annual revenue in a few months).

For venture capital in Web3 , the trend here is very different, as blockchain provides the perfect infrastructure layer for DeAI, such as: verifiable/untamperable transaction records, trustless environments, decentralized computing, trust-minimized AI reasoning and training (sorry, a lot of terms are used, but you should understand what I want to express).

In short, the future direction is to understand how their data is processed, understand the AI ​​thinking process, have their own data, have their own models and usage scenarios, and have the motivation to share (not censored), etc. Web3 venture capital has already invested in these future trends.

AI Agent Boom in Retail Market: More than Just Entertainment

DeAI is very difficult to understand for the Web3 retail market because it requires you to learn a lot of terms and understand the key points (sometimes it feels like an alien language). That's why retail market users prefer the easiest things to understand—such as the "Web3 AI Agents" that start with conversational bots, which are funny and can do entertaining content.

As the retail market continues to penetrate the industry, they gradually realize that these basic skills of chat and analytics are not enough to create sustainable value for users. This perception (plus the bad market environment) prompts the market to integrate, useless agents gradually disappear, while useful agents still survive (although their valuations are greatly reduced).

People are beginning to realize that AI products must have a core practical use scenario. This awareness prompted the team to either develop real AI products or work with truly technical AI companies, such as @AlloraNetwork , @opentensor (Bittensor).

There are two benefits to this transformation:

(i) It gives people a better understanding of infrastructure.

(ii) It provides real use scenarios for AI agents, allowing them to show their value to the community.

• Before the transformation: Agents with basic skills/use scenarios (chat, analysis, etc.)

• After the transformation: Agents with advanced practical skills (such as AI-driven betting, trading, liquidity provision, agriculture, etc.)

Agents like @AskBillyBets and @thedkingdao have become ideal agents, showing off the Bittensor subnet and bringing cool technology to the mainstream.

Bittensor Ecosystem: New Opportunities for Investment in Decentralized AI

I think one of the interesting things about the Bittensor ecosystem is that it is an ecosystem full of decentralized AI that anyone can invest in. Today, most decentralized AI projects are limited to VCs or strategic investors participating in a closed-door situation, as they are still in their early stages and many projects have not yet issued tokens.

But Bittensor allows anyone to stake their $TAO into the subnet they want to support, thus converting it into the subnet's alpha token (directly participating in the DeAI project).

Although I have publicly expressed disappointment with the bridging and trading experience, Bittensor’s technology, products and atmosphere are excellent, especially the team at @rayon_labs .

I like Rayon Labs because they do a lot of consumer-friendly work in optimizing UI/UX. Given the characteristics of dTAO – the market determines the emissions of each subnet and the pricing of the subnet – it becomes especially important for each subnet to build products that are easy to understand and understand.

Rayon has many cool subnets (the coolest of which is probably Gradients, an AutoML platform that allows for easy training of models on the platform), and even cooler is their latest flagship Squad AI proxy platform, which allows users to create proxy through drag and drop boxes (like the way that AI proxy is created in Figma).

Conclusion

I'm still in the early stages of getting to know Bittensor in depth and will be posting a dedicated post later to share interesting content I've found and show how to take the opportunity from it.