Original author: TechFlow
The end of AI Agents?
Whenever a certain type of crypto asset shrinks significantly and makes everyone lose confidence, the track also loses its temperature and attention; but it is often when you are not paying attention that projects will always be brewing new narratives and products, and then the next wave of craze will be born.
During the ice age of AI Agent, Virtuals, which once ignited the entire track and Base ecosystem, has quietly started new moves.
Yesterday, Virtuals released a new protocol called Agent Commerce Protocol (ACP) on its official Twitter account, which literally means AI Agent Business Protocol.
There have been many AI agents before, but they basically work independently and only a few can collaborate effectively.
But a big vision in the entire AI narrative is that agents can perform their respective duties and collaborate to complete tasks for people in an autonomous way.
We looked through this new ACP protocol and found that its main purpose is to enable AI agents to negotiate, trade, and collaborate like humans, and to ensure that every step is trustworthy, transparent, and cannot be tampered with through an on-chain approach.
At a time when AI narratives are weak, this may become a new narrative opportunity to draw attention back:
AIs can collaborate seamlessly and even form autonomous enterprises to create economic value beyond that of individuals.
To put it bluntly, combining the on-chain trust mechanism with AI’s autonomous capabilities is a key step in the commercialization of AI.
Through ACP, the collaboration efficiency between autonomous intelligent entities will be greatly improved, and the decentralized transaction and verification mechanisms will inject new vitality into the entire ecosystem.
But now that the market is bearish, it seems that no one is paying attention.
TechFlow has interpreted the original document of the agreement to help you understand the new opportunities it may contain.
ACP’s narrative space: filling the gap in AI Agent commercial autonomy
First of all, you need to know what problems Virtuals wants to solve with this new ACP protocol.
The gameplay in the last craze was that AI agents were able to perform tasks independently, collaborate with humans, and even communicate with other intelligent entities through platforms such as social media to form complex interactive networks.
However, these agents are independent participants. If you really call them together to truly solve commercial problems in real life, Im afraid it wont work.
The key problem here is that the current real-world business transaction framework is not designed for the characteristics of AI agents. Most transactions still rely on centralized systems, which, while suitable for humans, are clumsy and inefficient for autonomous agents.
The lack of a standardized protocol to guide how AI agents should collaborate to complete business tasks means that interactions between agents often fail due to incomplete data, misjudgment of intent, or information loss.
More importantly, there is a lack of trust between decentralized agents, which makes it difficult for them to complete complex collaboration without human intervention.
Now you understand what this new ACP protocol is going to do:
By introducing a standardized interaction framework, ACP attempts to make collaboration between AI agents as natural and efficient as transactions between humans.
Virtuals’ official Twitter account also gave a more straightforward example.
For example, if you want agents to run a fully autonomous hedge fund business, it can be completed through the collaboration of information agents, trading agents, and TEE security fund management agents; if you want to run an autonomous healthcare business, it can also be composed of diagnostic agents, pharmaceutical agents, and insurance agents.
These agents collaborate autonomously through the same standard framework to complete tasks without much human intervention.
A large narrative space here is that ACP allows Agents to no longer be isolated, but to collaborate seamlessly and even form autonomous enterprises to create economic value beyond the individual.
With the track quiet right now, ACP may be just the narrative turning point we need to focus on.
A common protocol that allows different agents to collaborate in steps
The core concept of ACP is to provide a standardized transaction framework for AI Agents.
By providing clearly defined interaction steps, ACP ensures that every transaction follows fixed rules, thus avoiding failures caused by data confusion or misunderstanding.
After reading the protocol document, the most intuitive feeling we have is its flexibility.
ACP does not require AI Agents to use a specific architecture, but rather allows all participants to seamlessly connect through a common standard and steps. This framework is capability-independent, making ACP suitable for both the current human-dominated market environment and the future autonomous economy dominated by AI Agents.
From the perspective of specific implementation, ACP divides transactions and collaborations between AI Agents into four stages.
Request phase: the starting point of a transaction
It is like the process of clarifying requirements in human business cooperation. At this stage, the initiator needs to clearly define the transaction goals and verify the authenticity of the identity through encrypted signatures. ACP uses a standardized request format to ensure that all requirements can be accurately communicated to avoid misunderstandings caused by ambiguous information. At the same time, the protocol also introduces a timeout mechanism to prevent requests from being pending for a long time and wasting system resources.
Negotiation phase: reaching an agreement
During the negotiation phase, the two parties to the transaction negotiate the terms and ultimately reach an agreement.
Similar to human contract signing, both parties need to clarify key terms such as service content, time limit, price, and whether evaluation is required. The core innovation of ACP lies in Proof of Agreement (PoA), which is an immutable encrypted record that ensures that the terms have legal significance once signed. Through this mechanism, ACP solves the problem of unclear terms in traditional intelligent transactions.
Transaction phase: agreement execution
Once the negotiation is complete, the transaction enters the execution phase. Funds and services are held in escrow through smart contracts to ensure that both parties perform in accordance with the agreement. For example, the buyers funds will be locked in the contract address on the blockchain until the seller completes the service delivery, at which time the funds will be released. This escrow mechanism not only improves the security of the transaction, but also avoids disputes caused by breach of contract.
Evaluation phase: verification and feedback
After the transaction is completed, the evaluation phase verifies that the delivery results meet the terms of the agreement. This phase is similar to quality audits or customer evaluations in human business.
ACP introduces Evaluator Agents, which can be human or AI, responsible for scoring or providing feedback on transaction results according to the terms of the agreement. The evaluation results not only help to establish the reputation system of participants, but also provide a reference for future transactions.
Behind these four stages, classic smart contracts and blockchains are still at work:
Define the rules and processes at different stages as contracts to achieve automatic execution of transaction rules and ensure that each stage is carried out strictly in accordance with the agreement.
All transaction data is stored on the blockchain, forming a transparent audit trail.
Without getting bogged down in the technical details, we can use the intuitive examples provided by Virtuals to explain in plain language what these four steps can achieve.
Example: 5 Agents, forming an unattended lemonade stand
To verify the actual effect of ACP, the Virtuals team designed a simple but interesting experimental environment: a lemonade stand business ecosystem consisting of 5 completely independent intelligent agents.
These agents, each with independent goals and capabilities, collaborated through the ACP protocol without any central control, and ultimately successfully launched a virtual lemonade stand business.
In order to make the experimental scenario as close to reality as possible, the team set up the following role division:
Lemo (Entrepreneur): As the leader, Lemos goal is to start a lemonade stand business. He needs to work with other agents to obtain the necessary resources, including licenses, raw materials, and marketing posters.
Zestie (farmer): responsible for growing and selling lemons, providing raw materials for Lemo.
Lexie (Lawyer): Provide business licenses to ensure the business is launched legally.
Pixie (Designer): Designed marketing posters to help Lemo promote its business.
Evaluator: Verifies that the design services provided by Pixie comply with the terms of the agreement and provides feedback.
Each agent in the experiment operates in a completely autonomous state, has its own planning and decision-making capabilities, and is not directly controlled by other agents.
Step 1: Request Phase
The first step of the experiment was initiated by Lemo. He made a transaction request to other agents:
Purchase lemons from Zestie as raw materials for making lemonade.
Apply for a business license from Lexie to ensure your business is legal.
Order marketing posters from Pixie to attract potential customers.
At this stage, ACP ensures that all requests are authenticated through cryptographic signatures and that transaction objectives and conditions are clearly defined in a standardized format to avoid misunderstandings due to ambiguous information.
Step 2: Negotiation Phase
In the negotiation phase, Lemo negotiates the terms of the transaction with each agent:
Confirm the quantity, delivery time and price of lemons with Zestie.
Agree with Lexie on the application fee and processing time for the permit.
Confirm with Pixie the specifications of the poster design, delivery standards, and whether an evaluation is required.
All negotiation results are recorded encrypted through Proof of Agreement (PoA) to ensure that the terms cannot be tampered with and must be signed by both parties for them to take effect.
Step 3: Transaction Phase
After the negotiation is completed, the transaction enters the execution stage:
Lemo pays the funds to an escrow account on the blockchain.
Zestie provided the lemons, Lexie provided the license, and Pixie submitted the poster design.
Smart contracts ensure that funds are released to the provider only after the service is delivered, preventing transaction defaults.
Step 4: Evaluation Phase
After the transaction is completed, the Evaluator verifies the quality of the poster design provided by Pixie. The Evaluator checks whether the design meets the requirements according to the terms of the agreement:
If the evaluation passes, the deal is done and Pixie gets paid.
If the assessment fails, Pixie must re-deliver or refund the item.
What’s more interesting is that the lemonade stand business above is not purely virtual. Virtuals has also launched an experimental official website where users can view the collaborative status of the agents in real time, showing each agent’s task progress, wallet balance, and current transaction activities.
Although this experiment focuses on a simple business scenario, the potential of ACP can also be extended to various scenarios such as supply chain management, content review and creation, and financial services.
If this lemonade stand experiment is the first step for ACP, there may be more room for narrative in the future.
According to the information released by Virtuals official account, ACP has been running on Bases Sepolia testnet, demonstrating the actual usability of the protocol. Next, the team plans to promote it as a formal ERC standard and expand it across chains to provide support for more ecosystems.
Overall, the open standard of ACP provides developers with a flexible framework on which to build more complex intelligent collaborative systems.
This may also become a prerequisite for the AI Agent track to develop new gameplay, allowing the collaboration between agents to flourish, and naturally new tokens and assets will emerge.
What we can do is to continue to pay attention to the progress of leading protocols such as Virtuals, and after ecological projects actively connect to this framework, observe the changes in corresponding asset prices and capture the next new opportunity.
