Redmond: Microsoft imagines a future where any company’s artificial intelligence agents can work with agents from other companies and have better memories of their interaction, said its main technologist on Sunday prior to the company’s annual software developer conference.

Microsoft will hold its Build conference in Seattle on May 19, with analysts expecting the company to reveal its latest tools to developers building AI systems.

When he spoke at Microsoft’s headquarters in Redmond, Washington, ahead of the conference, Chief Technology Officer Kevin Scott journalists and analysts told the company to focus on helping to the adoption of standards across the technology sector who will let agents from various manufacturers cooperate. Agents are AI systems that can perform specific tasks, such as solving a software error, on their own.

Scott said Microsoft supports a technology called Model Context Protocol (MCP), an open source protocol introduced by Google-supported anthropic. Scott said MCP has the potential to create an “agent web” that is similar to the way Hypertext protocols that helped spread the Internet in the 1990s.

“That means your imagination gets to run what the agent web becomes, not just a handful of companies that happen to see some of these problems first,” Scott said.

Scott also said that Microsoft is trying to help AI agents have better memories of things that users have asked them to do, noting that so far most of what we are building, very transactional. “

But making an AI agent’s memory better costs a lot of money because it requires more computing power. Microsoft focuses on a new approach called Structured Increase Increase In which an agent extracts short bits of each trip in a conversation with a user, creating a roadmap for what was discussed.

“This is a core part of how to train a biological brain – you don’t force to force everything in your head every time you need to solve a particular problem,” Scott said.