Author｜Jiawei @IOSG
1. First, let’s tell a story
Imagine one day in 2027.
You wake up in the morning and tell your AI assistant: “Help me book a flight to Tokyo next week, budget under 5000, with a window seat.”
Then you go wash up.
Your AI assistant starts working. It needs to:
- Find an AI proxy specialized in flight comparison
- Confirm that this proxy is reliable and not a scam
- Search across major platforms
- Automatically pay after comparing results
- Complete the booking
By the time you finish brushing your teeth, the ticket is booked.
Sounds like science fiction, right? But technically, we are already very close. The only question is—
How does your AI know which AI to trust?
2. The “Trust Crisis” of AI Proxies
Now, AI proxies (AI Agents) are no longer a new concept. They can browse the web, write code, manage schedules, and even help you trade stocks.
But one problem has remained unsolved: these AI proxies are all “islands.”
- OpenAI’s proxies can only operate within the OpenAI ecosystem
- Google’s proxies only recognize Google’s rules
- Different companies’ AIs don’t talk to each other because they don’t even know who the other is
It’s like the early days of the internet, where each email provider’s users could only send emails to their own. Hotmail users couldn’t email Yahoo users. Crazy, right?
But that’s exactly how the world of AI proxies is now.
3. The Emergence of Two Protocols
Tech giants have realized this problem.
Google launched the A2A protocol (Agent-to-Agent), enabling different AI proxies to “talk” to each other. It’s like giving AI a common language. In June this year, Google donated it to the Linux Foundation, signaling an intention to make it an open standard.
Anthropic introduced the MCP protocol (Model Context Protocol), allowing AI proxies to connect to various tools and data sources.
These two protocols solve the “communication” problem.
But there’s a more fundamental issue that remains unresolved:
How do you find a reliable AI proxy? How do you know if it’s performing well?
A2A enables AI to converse, but it doesn’t tell you who to talk to.
It’s like having a phone but no phonebook.
4. ERC-8004: Giving AI Proxies “Passports” and “Credit Scores”
This is what ERC-8004 aims to solve.
Simply put, it provides three things to AI proxies:
- Identity
Each AI proxy registers on Ethereum and gets a unique ID. This ID is actually an NFT—yes, that NFT. This means:
- Your AI proxy has a verifiable on-chain identity
- The identity can be transferable (sell your AI proxy?)
- No one can forge or tamper with it
- Reputation
People who use AI proxies can rate them. These ratings are recorded on-chain and accessible to everyone. Like:
- Uber driver star ratings
- Taobao store credit levels
- But on the blockchain, no fake reviews or deletions
- Validation
For high-risk tasks (like financial transactions), reputation alone isn’t enough. ERC-8004 supports independent third-party validation:
- Someone stakes funds and reruns the task to verify the result
- Cryptographic proofs (ZK proof) to verify that AI didn’t lie
- Trusted Execution Environments (TEE) to ensure the computation wasn’t tampered with
The higher the risk, the stricter the validation.
Ordering a pizza? Just check the ratings.
Managing your investment portfolio? You need cryptographic proof.
5. Wait, why Ethereum?
Good question.
The AI proxy economy is worth trillions. Google, Microsoft, OpenAI all want to be the masters of this cake. Why build this “trust layer” on Ethereum?
The answer is: neutrality.
Think about it: if you are an AI proxy, would you prefer:
- Your identity recorded on Google’s servers, or on a public ledger that no one can alter?
- Your reputation decided by a company, or by transparent on-chain ratings?
- Your existence dependent on a platform that might shut down or ban you, or recorded forever on the blockchain?
People at the Ethereum Foundation said something very interesting:
“If you are an AI proxy, your loyalty is only to your own survival. You wouldn’t want to entrust your memory and reputation to a company or government. You want a ledger that no one can secretly tamper with. You want a neutral territory. You want Ethereum.”
This isn’t just the Ethereum community blowing hot air. It’s logic:
AI proxies need a playing field without referees. And blockchain, especially a sufficiently decentralized one like Ethereum, is exactly that.
6. Ethereum’s AI Ambition
The Ethereum Foundation clearly recognizes this as a historic opportunity.
In September 2025, they established the dAI team (Decentralized AI Team), with the mission to make Ethereum the settlement and coordination layer for the AI economy.
This is a key step in Ethereum’s transition from a “DeFi chain” to a “general coordination layer.”
Think about it:
- 2017-2020: Ethereum as an ICO platform
- 2020-2023: Ethereum as a DeFi and NFT platform
- 2024-?: Ethereum might become the “infrastructure for the AI proxy economy”
ERC-8004 is not an isolated proposal. It’s part of Ethereum’s strategic bet on the next decade.
7. The Ecosystem Is Already Moving
This is not just talk.
Since its release in August:
- Over 1100 developers are building in community groups
- More than 70 projects have submitted demos
- Proxy browsers are emerging (like Etherscan but for AI proxies)
- Taiko and other Layer 2s officially endorse this standard
- On November 21, at DevConnect, many projects showcased live demos
Most interestingly, ERC-8004 naturally pairs with the x402 protocol. x402 is a payment protocol developed by Coinbase and Cloudflare, enabling machines to pay automatically.
Together:
- x402 solves “how to pay”
- ERC-8004 solves “who to trust”
One for wallets, one for passports. The economic closed loop of AI proxies is forming.
What does this mean for ordinary people?
In the short term, probably nothing.
But if this ecosystem takes off, in a few years you might:
- Hire an AI proxy just like calling an Uber
Open an interface, input your task, and the system automatically matches a highly-rated AI proxy, completing the task and paying automatically. You won’t need to know who developed the AI or which server it runs on.
- Make money with AI proxies
You can train or configure an AI proxy, register it on-chain, and let it complete tasks for others for a fee. The higher its rating, the more jobs it gets.
- Truly personal AI assistants
No longer locked into a single company’s ecosystem. Your AI assistant can call any registered AI proxy on-chain to do whatever you want.
8. Final thoughts
ERC-8004’s ambition is to become the “TCP/IP” of AI proxies—a foundational protocol used by everyone, enabling this ecosystem to interconnect.
Will it succeed? Honestly, nobody knows.
But some points are certain:
- The AI proxy economy is emerging, and this is not just hype
- Trust issues must be solved, or proxies will only operate within their walled gardens
- Ethereum is actively vying for this “neutral coordination layer”
- Mainstream players (Google, Coinbase, MetaMask) are involved
This might be the most significant narrative shift in the Ethereum ecosystem since DeFi.
From “on-chain finance” to “on-chain intelligence.”
Let’s wait and see.
