The Model Context Protocol (MCP) is one of the basic building blocks of AI interoperability, giving AI models a secure way to access external data sources and services. It’s the plumbing that lets a chatbot reach into your calendar, your database, or your internal tools, rather than of engineers building custom pipes for every connection. Next week, that protocol is getting a enormous update, and at the same time as it might not be noticeable to end users, it is able to make a huge difference in how the ecosystem develops.
The official spec for the latest version has been public since May, however we got an unusually clear explanation of the changes Monday morning from the folks at Arcade — a two-year-old startup that’s built its complete business around the work of getting AI agents to clearly function inside real companies, permitting them to securely link to and act on tool like Gmail, Slack, and Salesforce.
Arcade raised $60 million in June based on the idea that most AI agents don’t fail because the models are weak however because the infrastructure around them isn’t prepared yet, and that’s what this update is trying to cope with. Importantly, MCP is changing the way it handles session IDs — the little tokens that servers use to remember “ah, this is the same conversation as five seconds ago” — so servers can operate more easily at a larger scale.
As Arcade founding engineer Nate Barbettini puts it:
[Under the current system] The first time an MCP client like Claude links to a server, it sends a “hello”: I’m Claude, right here’s my version, here are my capabilities. The server replies with its own capabilities and hands a session ID… From then on, the client sends that session ID on every request so the server knows it’s the same conversation. Sometimes the ID expires, so the client has to notice, request a new one, and carry on….
Picture a real deployment. You’re running a server for millions users, behind a load balancer whose complete job is to route each request to whatever server within the farm is free, sometimes in a different region. Now each one of those machines has to know about a session ID that a some other device handed out. It’s not impossible, but it’s a serious pain, and it fights the load balancer rather than of working with it.
In different phrases, the recent setup assumes one server remembers you, however real corporations spread traffic across dozens of servers that don’t talk to each other by default, so today’s MCP servers have to do extra work simply to maintain track of who’s who. That’s been a huge headache for anyone of us running an MCP server at scale, and part of the reason we haven’t seen more companies ship large-scale, first-party MCP incorporations instead all the hype around agentic AI this year.
Under the new system, the protocol will take a looser, “stateless” technique to session IDs on the server aspect, similar to how most ordinary websites already work, which have to make the whole system lots easier to preserve and, in theory, cheaper to run at scale.
That’s all pretty technical, however it’s an essential reminder that now not every part of AI development is moving at breakneck speeds. While model training races ahead, a lot of the technical infrastructure those models require is still subject to be slow log-rolling of standards-body consensus. It really is happening; it’s just a little slower!
