Microsoft 365 Copilot will add the ability to run OpenAI and Anthropic AI models simultaneously in a single prompt.



Microsoft's AI service, '

Microsoft 365 Copilot ,' has begun testing a feature in its Researcher function that allows users to run two AI models simultaneously to produce high-quality reports.

Introducing multi-model intelligence in Researcher | Microsoft Community Hub
https://techcommunity.microsoft.com/blog/microsoft365copilotblog/introducing-multi-model-intelligence-in-researcher/4506011

Copilot's Researcher is a feature that can create detailed reports based on internal company data such as emails, chat histories, meeting recordings, and documents. With Researcher, you can choose to run either OpenAI's GPT or Anthropic's Claude. This update adds an 'Auto' option to the model selection screen. Auto is a mode that uses two types of models in combination, with one handling 'draft creation' and the other performing 'review and refinement.' Microsoft has named the Auto mode feature 'Critique.'



The graph below compares the performance of Copilot's Auto mode with OpenAI Deep Research, Gemini Deep Research, Claude Opus 4.6, and Perplexity Deep Research. It shows that Auto mode can analyze information with very high accuracy.



Additionally, a new mode called 'Council' has been added, which allows you to run two different models simultaneously for a single input.



The output in Council mode is shown below. The GPT and Claude generation results are output in a comparable format.



Critique and Council are open to participants of the Microsoft 365 Copilot Frontier program .

Related Posts:

in AI, Posted by log1o_hf