Microsoft 365 Copilot gets Critique and Council modes

Microsoft added Critique and Council, two multi-model capabilities, to Microsoft 365 Copilot Researcher.

Critique uses a two-model draft-and-review workflow and scored 13.88% higher than Claude Opus 4.6 on the DRACO benchmark.

Council runs Anthropic and OpenAI models side-by-side and provides a judge-model summary of consensus and differences.

Both features are available to organizations enrolled in the Microsoft 365 Copilot Frontier program in supported regions.