AI core model provider Anthropic has proposed a new framework for connecting AI assistants to their data.
The generative AI provider, developer of the Claude family of large language models and rival to OpenAI, introduced the Model Context Protocol on November 25. Anthropic said MCP provides a universal open standard for connecting AI systems to data sources.
MCP allows developers to expose their data through MCP servers or create AI applications that connect to the servers. The three main components of MCP are the Model Context Protocol specification and SDKs, local MCP server support in Claude desktop applications, and an open source repository of MCP servers.
Anthropic said MCP is open source and aims to help frontier models produce better answers.
Anthropic’s introduction of MCP comes as agentic AI continues to gain traction in the AI market and more vendors introduce autonomous and semi-autonomous AI agent technology. With agentic AI and LLMs, developers need to streamline the process of connecting data to their AI agents or assistants.
More suppliers needed
However, said Mark Beccue, an analyst in TechTarget’s corporate strategy group, introducing a standard is not something a standalone vendor can do.
A standard is when the community agrees that this is what we’re going to do, and it’s collaborative.
Marc BeccueAnalyst, Corporate Strategy Group
“A standard is when the community agrees that this is what we’re going to do, and it’s collaborative,” Beccue said. “They opened this up, but they’re the only ones talking about it. I just think it’s a failure.”
He added that Anthropic could explore with MCP a standardized way for LLMs to communicate with proprietary data. Anthropic may then need to start a conversation with other suppliers about what standard is needed.
However, a standard would be better if it came from a group like the AI Alliance, which was started by IBM, Meta and other AI vendors, Beccue argued. Indeed, it would be the decision of a group of suppliers rather than just one.
Anthropic said in a blog post that it is committed to making MCP a collaborative open source project and that it seeks feedback.
” Unlike other AI labs, Anthropic does not train our generative AI models on user-submitted data by default.“, according to an Anthropic spokesperson.
Developers and the data question
Still, MCP is useful for developers looking for a streamlined process for connecting AI models to their data, said Sean Ren, associate professor of computer science at the University of Southern California and CEO of Sahara AI , provider of an AI blockchain. platform.
“(Anthropic) allows you to follow the protocol to connect various resources into large models and apply tools on top to realize the application,” Ren said. “It looks more like a frame.”
Although MCP can help developers, Anthropic has not explained how it will protect user privacy and data security, Ren said. He noted that even though MCP is open source, users still access it through Anthropic’s closed model, Claude.
“Individual companies need to be very careful when using the protocol, because you are using it to connect your personal data to Anthropic models,” he said. “Essentially, you’re giving them access to your very private data, and there are no clear details about how they’re going to protect that data, or whether they’re going to access it for other uses.”
Anthropic also revealed on November 26 that users of Claude.ai, the web version of the LLM, can now tailor their answers based on their preferences, tone and structure. Users can choose from these styles: formal, concise, and explanatory.
Esther Ajao is a news editor at TechTarget and host of podcasts covering artificial intelligence software and systems.
