AI ushers in the "USB-C moment"; how will MCP perfectly integrate with Ethereum?

Content | Bruce

Editing & Typesetting | Huanhuan

Design | Daisy

The “USB-C moment” in the evolution of AI is here. In November 2024, the MCP protocol released by Anthropic is causing a seismic shift in Silicon Valley. This open standard, dubbed the “USB-C of the AI world,” not only redefines the connection between large models and the physical world but also conceals the key to breaking the AI monopoly dilemma and reconstructing the production relationships of digital civilization. While we are still debating the parameter scale of GPT-5, MCP has quietly paved the decentralized path to the AGI era…

Bruce: Recently researching the Model Context Protocol (MCP). This is the second thing that excites me very much in the AI field after ChatGPT, as it has the potential to solve three questions I have been pondering for years:

• How can non-scientists and ordinary people participate in the AI industry and earn income?
• What are the win-win combinations of AI and Ethereum?
• How to achieve AI d/acc? Avoid the monopoly and censorship of large centralized companies, AGI destroying humanity?

01. What is MCP?

MCP is an open standard framework that simplifies the integration of LLMs with external data sources and tools. If we compare LLMs to the Windows operating system, applications like Cursor are the keyboard and hardware, while MCP is the USB interface that supports the flexible insertion of external data and tools, allowing users to access and utilize these external data and tools.

MCP provides three capabilities to extend LLM:

• Resources
• Tools (execute functions, call external systems)
• Prompts

MCP can be developed and hosted by anyone, provided as a server, and can be taken offline to stop service at any time.

02. Why is MCP needed?

Currently, LLM uses as much data as possible for extensive computations and generates a large number of parameters, incorporating knowledge into the model to achieve dialogue outputs of relevant knowledge. However, there are several significant problems:

1. Large amounts of data and computations require a lot of time and hardware, and the knowledge used for training is often outdated.
2. Models with a large number of parameters are difficult to deploy and use on local devices, but in reality, most users may not need all the information to meet their requirements.
3. Some models use crawlers to read external information for computations to achieve timeliness, but due to the limitations of crawlers and the quality of external data, they may produce more misleading content.
4. Since AI has not brought significant benefits to creators, many websites and content are starting to implement anti-AI measures, generating大量垃圾信息, which will lead to a gradual decline in the quality of LLM.
5. LLM is difficult to extend to various external functions and operations, such as accurately calling the GitHub API to perform certain actions. It may generate code based on possibly outdated documentation, but it cannot ensure precise execution.

03, the architectural evolution of Fat LLM and Thin LLM + MCP

We can consider the current large-scale models as fat LLMs, whose architecture can be represented by the simple diagram below:

After the user inputs information, it is decomposed and reasoned through the Perception & Reasoning layer, and then a large number of parameters are called to generate results.

After the MCP, LLM may focus on language parsing itself, stripping away knowledge and capabilities, becoming a lean LLM:

Under the architecture of Slim LLM, the Perception & Reasoning layer will focus on how to parse comprehensive human physical environment information into tokens, including but not limited to: voice, tone, smell, image, text, gravity, temperature, etc. Then, tasks will be completed by orchestrating and coordinating up to hundreds of MCP Servers through the MCP Coordinator. The training cost and speed of Slim LLM will be greatly enhanced, and the requirements for deployment devices will become very low.

04. How MCP Solves Three Major Problems

How can ordinary people participate in the AI industry?

Anyone with unique talents can create their own MCP Server to provide services to LLM. For example, a birdwatcher can offer their years of bird notes through MCP. When someone uses LLM to search for information related to birds, it will call upon the current bird notes MCP service. Creators will also earn revenue share as a result.

This is a more precise and automated creator economy cycle, where the service content is more standardized, and the number of calls and output tokens can be accurately counted. LLM providers can even simultaneously call multiple Bird Note MCP Servers to allow users to select and rate to determine whose quality is better and obtain higher matching weights.

The Win-Win Combination of AI and Ethereum

a. We can build an OpenMCP.Network creator incentive network based on Ethereum. The MCP Server needs to host and provide stable services, users pay LLM providers, and LLM providers distribute actual incentives to the invoked MCP Servers over the network to maintain the sustainability and stability of the entire network, stimulating MCP creators to continuously create and provide high-quality content. This network will require the use of smart contracts to achieve automation, transparency, trustworthiness, and censorship resistance in incentives. Signature, permission verification, and privacy protection during operation can be implemented using Ethereum wallets, ZK, and other technologies.

b. Develop MCP Servers related to operations on the Ethereum chain, such as AA wallet invocation services, allowing users to make wallet payments through language in LLM without exposing related private keys and permissions to LLM.

c. There are also various developer tools that further simplify Ethereum smart contract development and code generation.

Achieve AI Decentralization

a. MCP Servers decentralize the knowledge and capabilities of AI, allowing anyone to create and host MCP Servers. After registering on platforms like OpenMCP.Network, they can receive incentives based on calls made. No single company can dominate all MCP Servers. If an LLM provider offers unfair incentives to MCP Servers, creators will support blocking that company, and users will switch to other LLM providers for a more equitable competition if they do not receive quality results.

b. Creators can implement fine-grained access control over their MCP Servers to protect privacy and copyright. Lean LLM providers should incentivize creators to contribute high-quality MCP Servers by offering reasonable incentives.

c. The gap in the capabilities of slim LLMs will slowly be bridged, as human language has a limit to its traversal and the evolution is also very slow. LLM providers will need to focus their attention and funding on high-quality MCP Servers, rather than repeatedly using more GPUs for mining.

d. The capabilities of AGI will be decentralized and downgraded, with LLM only serving as language processing and user interaction, while specific capabilities are distributed across various MCP Servers. AGI will not pose a threat to humanity, as after shutting down the MCP Servers, only basic language conversations can take place.

05. Overall Review

1. The architectural evolution of LLM + MCP Servers essentially decentralizes AI capabilities, reducing the risk of AGI destroying humanity.
2. The usage of LLM allows for token-level statistics and automation of the number of calls to MCP Servers as well as input and output, laying the foundation for the establishment of the AI creator economy system.
3. A good economic system can drive creators to actively contribute high-quality MCP Servers, thereby promoting the development of all humanity and achieving a positive feedback loop. Creators no longer resist AI, and AI will also provide more jobs and income, reasonably distributing the profits of monopolistic business companies like OpenAI.
4. This economic system, combined with its characteristics and the needs of its creators, is very suitable for implementation based on Ethereum.

06. Future Outlook: The Next Step in Script Evolution

1. MCP or similar MCP protocols will emerge one after another, and several large companies will start to compete in defining the standards.
2. MCP Based LLM will emerge, focusing on parsing and processing human language with small models, accompanied by MCP Coordinator connecting to the MCP network. LLM will support automatic discovery and scheduling of MCP Servers without the need for complex manual configuration.
3. MCP Network service providers will emerge, each with its own economic incentive system, and MCP creators can register and host their own servers to generate income.
4. If the economic incentive system of the MCP Network is built on Ethereum using smart contracts, then the transactions on the Ethereum network are conservatively estimated to increase by about 150 times (based on a very conservative daily call volume of 100 million MCP Servers, currently calculating 100 txs per block every 12 seconds).