Microsoft Brings New AI Capabilities to Copilot Researcher

Share This Post

Microsoft released two new capabilities for Researcher, its AI research agent for enterprises within its Copilot platform.

The updates, dubbed Critique and Council, are designed to improve the agent’s accuracy and “depth,” as Microsoft seeks to bolster the reliability of AI-generated research for enterprises.

In a blog post on the launch on Monday, Microsoft said the features better an agent’s ability to handle complex research tasks, combining multiple AI models and using the same evaluation processes typically seen in academic and professional research. 

Critique works by dividing the research process between two AI models. One is used for planning, retrieval and drafting, while the second evaluates accuracy, strengthens arguments and refines the final report.

By using two AI workflows (rather than a single model), Critique is designed to boost source reliability and transparency.

In performance assessments, the system was found to surpass single-model approaches across categories including accuracy, analytical breadth and presentation quality, the vendor said.

Related:OpenAI Now Valued at $852B After Funding Round

“Critique pushes Researcher to identify missing analytical angles, close coverage gaps, sharpen formulations, and produce responses with stronger organization and clearer narrative flow,” Microsoft said in the blog. “This accounts for the substantial improvements to the breadth and depth and presentation quality scores.”

Council was also designed to improve reliability, offering a side-by-side comparison of outputs from multiple AI models.

When activated, Council runs models from Anthropic and OpenAI in parallel, with each generating a complete, independent report. A separate “judge” model then evaluates both outputs to produce a summary highlighting key areas of agreement and divergence, along with any particular points of note.

Microsoft said the feature is designed to give users greater transparency into how different AI systems approach the same problem — an increasingly important consideration as companies deploy multi-model systems.

Both models are now available in Microsoft’s Frontier program.

Related Posts

Oil trader takes $17 million hit as tokenized crude rivals bitcoin liquidations

Crypto's biggest liquidation event this week wasn't about crypto.Tokenized...

Crypto Scam Leader Extradited to China to Face Charges

Li Xiong, a key member of a group that...

Strategy’s STRC maintains dividend at 11.5% after steady increases

Strategy, the world’s largest publicly traded Bitcoin holder, has...

Bitcoin slides to $66,600 as Trump threatens to hit Iran ‘extremely hard’

Bitcoin fell 2.2% to $66,609 on Wednesday, giving back...

Crypto asset manager CoinShares (CSHR) to list on Nasdaq after $1.2 billion SPAC deal

CoinShares, a leading European digital asset manager with over...