Episode

Architecture Over Chat: Building the Agent Fabric

Podcast
M365.FM - Modern work, security, and productivity with Microsoft 365
Published
Jun 28, 2026
Duration seconds
4708
Processing state
not_requested
Canonical source
https://www.spreaker.com/episode/architecture-over-chat-building-the-agent-fabric--72560443
Audio
https://dts.podtrac.com/redirect.mp3/api.spreaker.com/download/episode/72560443/architecture_over_chat_building_the_agent_fabric.mp3
JSON
/v1/public/podcasts/m365-fm-modern-work-security-and-productivity-with-microsoft-365-7311214/episodes/architecture-over-chat-building-the-agent-fabric
Markdown
/podcast/m365-fm-modern-work-security-and-productivity-with-microsoft-365-7311214/architecture-over-chat-building-the-agent-fabric.md

Actions

  • POST https://stenobird.com/v1/public/podcasts/m365-fm-modern-work-security-and-productivity-with-microsoft-365-7311214/episodes/architecture-over-chat-building-the-agent-fabric/transcription-requests
    Idempotently request low-priority transcript generation for this episode.
  • GET https://stenobird.com/podcast/m365-fm-modern-work-security-and-productivity-with-microsoft-365-7311214/architecture-over-chat-building-the-agent-fabric.md
    Read the agent-friendly Markdown representation of this episode resource.

Summary

Most organizations believe they are building AI agents. In reality, they are building chatbots trapped inside applications. These systems can answer questions and generate content, but they forget everything when a session ends. They cannot coordinate across systems, maintain long-term context, or operate as true workforce participants. In this episode, we explore one of the biggest architectural shifts happening in enterprise AI today: the move from isolated conversational experiences to persistent agent fabrics. Instead of treating AI as a chatbot inside Teams, Slack, or a web application, organizations must begin thinking about agents as long-running, governed, identity-driven participants that can operate across devices, applications, and business processes. The discussion examines why the problem isn't the intelligence of modern models. The real limitation is the infrastructure surrounding them. Memory, identity, governance, orchestration, observability, interoperability, and security have become the critical building blocks for the next generation of enterprise AI systems. THE CHATBOX ILLUSION Most AI deployments today are still built around conversations. While chat interfaces are familiar and easy to adopt, they create significant limitations when organizations attempt to scale AI beyond simple question-and-answer scenarios. Key topics include: Why chat is the wrong abstraction for enterprise agents The limitations of stateless architectures Why agents need persistent memory The difference between assistants and workforce participants BREAKING DOWN THE SILO PROBLEM Organizations are creating AI capabilities inside CRM systems, project management tools, customer service platforms, and productivity applications. Unfortunately, these agents often operate independe…