Microsoft is looking towards a future where AI agents, capable of performing specific tasks independently, can seamlessly collaborate across different companies. According to Chief Technolo…
Microsoft is looking towards a future where AI agents, capable of performing specific tasks independently, can seamlessly collaborate across different companies. According to Chief Technology Officer Kevin Scott, the company is prioritizing the development of industry-wide standards to facilitate this interoperability.
This vision involves an "agentic web" where AI agents from various developers can work together, much like the interconnectedness of the internet. Microsoft is supporting the Model Context Protocol (MCP), an open-source protocol from Anthropic, to achieve this vision. A key aspect of Microsoft's strategy involves enhancing the memory capabilities of AI agents.
Currently, Scott notes that many AI interactions feel "transactional." To address this, Microsoft is exploring techniques to enable agents to retain and utilize past interactions more effectively. This includes a new approach called structured retrieval augmentation, which involves extracting key information from conversations to create a roadmap of the discussion.
This method aims to improve efficiency and reduce the computational costs associated with enhancing AI memory. The company recognizes the significant computing power required to improve AI memory. Microsoft's structured retrieval augmentation strategy aims to make agents more efficient and cost-effective.
This approach is inspired by how biological brains function, focusing on targeted information retrieval rather than brute-force memorization. This will enable AI agents to build upon past interactions and provide more contextually relevant responses. Microsoft's focus on AI agent collaboration and improved memory reflects its broader commitment to advancing AI technology.
The company's initiatives at the Build conference and its backing of open-source protocols like MCP underscore its dedication to fostering a more open and collaborative AI ecosystem. By improving the way AI agents interact and remember, Microsoft aims to unlock new possibilities and drive innovation in the field.