Microsoft's Copilot research tool has been upgraded to use responses from OpenAI's ChatGPT and Anthropic's Claude simultaneously. The new Critique feature refines ChatGPT answers using Claude, creating a feedback loop aimed at improving factual accuracy, analytical breadth, and presentation quality. Microsoft claims this combined approach outperforms standalone models like Perplexity's Deep Research on key metrics.

This follows Anthropic's own multi-agent research system. For users preferring independent model comparisons, Model Council offers side-by-side results from both AIs with a disagreement report. Both features launch in Microsoft 365 Copilot's Frontier early access program.

This integration represents a significant step towards more powerful, multi-model AI research assistants.