Original: Deep Tide TechFlow

On January 28, the Ethereum official announced that the ERC-8004 protocol will soon be launched on the mainnet.

This standard was mentioned in our article last October. If you’re not familiar with it, you can refer to this: “x402 Gradually Involution, Preemptively Exploring New Asset Opportunities in ERC-8004”

It actually has a formal name called “Trustless Agents.” Translated into plain language, it roughly means:

Issuing on-chain IDs for AI Agents.

The Ethereum Foundation rarely pushes an ERC standard so vigorously. They specifically established a team called dAI, included ERC-8004 in the 2026 strategic roadmap, collaborated with Google, Coinbase, and MetaMask to draft proposals, and held a Trustless Agents Day at DevConnect in November to generate hype.

The last time Ethereum pushed a standard this earnestly was for ERC-20 and ERC-721.

One defined tokens, and the other defined NFTs.

Now it’s AI’s turn?

Ethereum’s AI Anxiety

Why the urgency?

Take a look at some data. According to statistics from Cookie.fun, the market value distribution of AI Agent tokens shows that Solana and Base combined account for 96%. There are only a handful of AI Agent projects on the Ethereum mainnet that are recognizable.

There are only a handful of AI Agent projects on the Ethereum mainnet that are recognizable.

In April 2025, the exchange rate of ETH to BTC fell to 0.017, a five-year low. At that time, everyone said Ethereum was not the future.

When DeFi was hot, Ethereum was the main stage. When NFTs were hot, Ethereum was also the main stage. Now that AI Agents are getting hot, the main stage has changed hands.

Solana processes 36 million transactions a day, while Ethereum’s mainnet handles 1.13 million transactions. High gas fees and slow speeds make developers vote with their feet. Virtuals Protocol launched on Base, and the previous ai16z chose Solana; even Coinbase’s own AI project was not placed on the Ethereum mainnet.

Ethereum needs a new narrative.

ERC-8004 might just be the beginning of that narrative.

Let’s Review ERC-8004

Let’s get back to the standard itself.

How exactly does ERC-8004 issue on-chain IDs for AI Agents?

You don’t need to understand any technology; just know about three registries.

The first is the Identity Registry. Based on ERC-721, each AI Agent mints an NFT to prove “I am who I am.”

The second is the Reputation Registry. It records the historical performance of this Agent, who has used it, what the evaluations were, and if it has committed any misdeeds.

The third is the Validation Registry. It allows third-party organizations to endorse the Agent, for example, “This Agent passed a certain security audit.”

Together, these three registries solve one problem: how do two AI Agents know if each other is trustworthy when they meet on-chain?

The previous answer was that we don’t know; we can only rely on humans. The answer from ERC-8004 is to check on-chain records.

This framework was not invented by Ethereum itself.

Its underlying logic comes from Google’s A2A protocol released last year, Agent-to-Agent, enabling AI to communicate and call each other. ERC-8004 adds an additional layer to this:

Trust backed by blockchain.

Google’s A2A addresses communication issues, while Ethereum’s ERC-8004 addresses trust issues. One handles communication, and the other verifies identity.

Is Issuing IDs a Good Business?

Let’s boldly speculate that Ethereum’s logic might be as follows:

For AI Agents to be truly useful, they need to manage money themselves. Not just tweeting or chatting, but directly operating on-chain assets. Signing transactions, adjusting contracts, cross-protocol arbitrage…

Currently, no one dares to do this on a large scale. The reason is simple: how can you be sure this Agent won’t transfer your money away? Recently, ClawdBot has become very popular, and community members have already posted related negative events.

The solution in Web2 is platform endorsement. You use OpenAI’s API, and trust comes from OpenAI. If something goes wrong, you turn to OpenAI.

Web3 doesn’t have this. Agents are open-source, deployment is permissionless, and no one supervises them on-chain. When you call a service from a stranger Agent, you can’t check who’s behind it, whether the code has issues, or if it has a history of wrongdoing… none of this information is accessible.

In simple terms, ERC-8004 essentially moves the traditional financial KYC process onto the chain. Ethereum bets that when AI Agents start dealing with real money, this framework will become a necessity.

DeFi protocols need to connect with external Agents, and they must first check their on-chain identity. Institutions looking to use Agents for trade execution must check their historical records. Audit firms can issue on-chain certifications for Agents, just like performing security audits for smart contracts.

This is a strategic positioning move in the competition.

Ethereum realizes it has already lost on the execution layer, but no one has occupied the trust layer yet. Institutional recognition, security audit ecosystems, and TVL scale are the assets Ethereum currently has. ERC-8004 packages these assets into a standard, defining “what compliance looks like for AI Agents” before anyone else.

The question is, does this demand exist now?

Standards Before Demand

Having discussed Ethereum’s plan, let’s talk about reality. What are AI Agents on the chain doing now?

With the end of last year’s AI meme wave and the rapid advancements of several leading AI companies’ products over the past year or two, there aren’t many people paying attention to on-chain AI Agents now.

However, they are still making progress.

For instance, ai16z has renamed itself ElizaOS, transforming from a single Agent into a cross-chain platform; Virtuals Protocol is working on AI DApps and plans to enter physical robotics in 2026; and other AI Agents like those in Surf can also automatically execute DeFi trading strategies.

But here’s the problem: do they really need ERC-8004?

Users trust Luna because it is developed by the core team of Virtuals. Agents on ElizaOS are used because they operate within the framework of ElizaOS; Surf helps execute your strategies, and often it’s the application itself that you trust.

Trust comes from the platform, not from on-chain identity.

The scenario envisioned by ERC-8004 is that a stranger Agent approaches you without platform endorsement or brand recognition, and you can only judge its trustworthiness through on-chain records.

When will this scenario occur?

When AI Agents can truly achieve autonomous calls across protocols, platforms, and organizational boundaries. An Agent borrows money from Aave, trades on Uniswap, and then goes to another protocol to yield, all without human approval…

But this scenario doesn’t exist right now.

Currently, AI Agents, no matter how complex their functions, essentially still operate within a single platform. They don’t need to prove themselves to stranger protocols because they don’t even knock on the doors of unfamiliar protocols.

Given the current heat of the crypto market, they also have no reason to knock on each other’s doors unless they can collectively create a new narrative.

Therefore, ERC-8004 addresses a future problem.

If AI Agents transition from toys to tools, Ethereum’s trust infrastructure will become valuable. If the scale of the Agent economy is large enough and cross-platform calls become the norm, ERC-8004 could start collecting tolls.

There are many “ifs.”

Thus, this forward-looking layout might be initiated first by institutions.

By the end of 2025, SharpLink Gaming announced an investment of $170 million into Ethereum’s restaking strategy. During the same period, over 23,000 ETH were net withdrawn from exchanges, flowing into private wallets and staking protocols.

This money might be buying Ethereum for 12 to 18 months down the line.

For retail investors, ERC-8004 isn’t exactly a great catalyst.

Betting on ERC-8004 itself? It’s an open standard, has no tokens, and can’t be directly invested in; you can only look for some related small projects. It’s also possible to bet on Ethereum, but Ethereum’s price is influenced by too many factors, and AI Agents are just one narrative among them.

Therefore, there is currently no clean asset that allows you to precisely bet on the proposition that “AI Agents need on-chain identities.”

Ethereum isn’t entirely the infrastructure for AI, and its identity anxiety won’t be completely alleviated by AI. The business of creating AI IDs still has a long way to go.

Welcome to join the official ChainCatcher community

**_Telegram Subscription:

**_Official Twitter Account: