Search

An inside look at the data science and evaluation systems helping teams improve agent quality at scale in Copilot Studio.

Frontier Tuning introduces a new way to tune AI agents around your organization’s workflows, internal knowledge, and compliance safeguards.

Copilot Studio adds Mistral Medium 3.5, expanding model choice with in‑region data control, strong multilingual performance, and admin governance.

Learn what’s new in Copilot Studio, May 2026: computer-using agents are now generally available, plus redesigned workflows and Work IQ extensibility.

Explore an in-depth guide to managing customer-facing real-time voice agents with Copilot Studio, from governance foundations to production readiness.

See what’s new in Copilot Studio, April 2026: updates to workflows, increased control over agent operations, and an expanded agent usage estimator.

Real-time voice agents are now generally available in Copilot Studio, supporting adaptive voice experiences for complex customer conversations.

Introducing new capabilities in Microsoft Copilot Studio that help you automate your business processes by mixing AI agents and workflows.

Learn what’s new in Copilot Studio: Multi-agent systems are now generally available, plus recent updates to the Prompt Editor and governance controls.

Agentic AI introduces new security risks.

Custom Graders in Copilot Studio close the gap between what “correctness” measures and what organizations need from their agent evals.

Wave 3 marks a new version of Microsoft 365 Copilot, moving beyond assistance to embedded agentic capabilities.