Microsoft wants its AI ‘bots’ to collaborate and retain information.
Ahead of the company’s annual software developer conference, Microsoft’s top technologist stated on Sunday that the company sees a future in which its artificial intelligence agents can collaborate with agents from other companies and have better recollections of previous interactions.
Analysts anticipate that Microsoft will introduce its newest tools for developers creating AI systems at its Build conference in Seattle on May 19.
Prior to the conference, Chief Technology Officer Kevin Scott told reporters and analysts at Microsoft’s Redmond, Washington, headquarters that the company is committed to promoting the adoption of industry-wide standards that will enable agents from various manufacturers to work together. AI systems known as agents are capable of carrying out particular tasks, such resolving a software bug, by themselves.
According to Scott, Google-backed Anthropic introduced the open-source Model Context Protocol (MCP) technology, which Microsoft is supporting. According to Scott, MCP might produce a “agentic web” in a manner akin to the hypertext protocols that facilitated the internet’s growth in the 1990s.
“It means that your imagination gets to drive what the agentic web becomes, not just a handful of companies that happen to see some of these problems first,” Scott stated.
According to Scott, Microsoft is working to improve AI agents’ recall of tasks that customers have requested of them, but as of right now, “most of what we’re building feels very transactional.”
However, improving the memory of an AI agent is expensive because it necessitates more processing power. Microsoft is concentrating on a novel strategy known as structured retrieval augmentation, in which an agent summarizes each turn in a user discussion to provide a roadmap of the topics covered.
“This is a core part of how you train a biological brain – you don’t brute force everything in your head every time you need to solve a particular problem,” Scott stated.
