A conversation with Charles Lamanna on building custom copilots with Microsoft
summary
Microsoft demos building custom copilots with Copilot Studio that wires together content labeled as knowledge, editable instructions, and actions (workflows/APIs) so the LLM decides when to call each piece. Agents run disconnected from chat to respond to events or APIs and run in the background. AI recorder captures UI actions to create robust desktop automations; AI flows describe outcomes and dynamically assemble cloud workflows. Claim: enterprises will have more copilots than people; example: a holiday-shopping copilot built in under three minutes. Security ties into Defender and Sentinel audit events.
AI-written summary · about
chapters (9)
- 0:00Microsoft's AI vision
- 0:52Understanding AI agents
- 2:08Agents in action
- 3:41Co-pilot studio
- 5:05Future of development
- 5:50AI recorder for RPA
- 7:26Dynamic AI flows
- 8:44AI security measures
- 9:51Getting started
related talks
49mBuild your custom Copilot with Microsoft Copilot Studio
45mThe era of the AI Copilot
27mLarge-scale Enterprise Co-pilots - Making AI Useful for Business
23mKEYNOTE: From Copilot to Flight Captain
45mExploring Copilot Capabilities
56mBuild your own CoPilot
43mHarness the power of AI: Extend Copilot and beyond
46mHow work gets done in the Age of Al: Copilot, Apps and Agents