Editorial illustration for Microsoft adds Critique and Council to Copilot following USD 1B Disney blindside
Microsoft Copilot Gains Critique & Council Features
Microsoft adds Critique and Council to Copilot following USD 1B Disney blindside
Microsoft just made a quiet power play that changes the calculus on closed platforms. The company’s Copilot Researcher, already built on OpenAI’s models, will now also call on Claude to act as a built-in critic, tearing apart its own work before you ever see it. That means one AI drafts, another audits.
Two systems run side by side, showing where they agree and dissent. The move comes on the heels of OpenAI’s staggering $1 billion deal with Disney, a blindside that made clear how quickly partnerships can shift. Now Microsoft has stepped back from a single-model dependency, turning its own tool into a multi-model battleground.
The trade-off is blunt: you trade simplicity for rigor, convenience for cross-checking. And if you’re planning your next model switch, this changes the math.
The strangest part of the story is the Disney blindside, which is certainly a strange way to handle a potential $1B partnership with one of the biggest media companies on the planet.
Microsoft just turned Copilot into a courtroom. One model builds the case. Another cross-examines it.
That’s not a feature update, that’s a quiet admission: no single AI is trustworthy enough to go it alone. The Disney blindside, a billion-dollar lesson in the cost of blind trust, makes that official. So what do you, the enterprise buyer, do with this?
You stop chasing the next model switch as a silver bullet. You start demanding systems that argue with themselves. Because the winner in this race won’t be the model with the highest benchmark.
It’ll be the platform that institutionalizes doubt, and still ships.
Related Reading
Browse all Microsoft news →
Market Trends AI drives three initiatives to accelerate global action for planet protection
Market Trends Akamai data shows AI training bots and content‑fetching bots rise since July
Market Trends NVIDIA and Google Cloud let developers scale AI from prototype to production
Related Topic Claude gains shared context in Excel, PowerPoint; Microsoft adds Copilot Cowork
AI Tools & Apps
Related Topic Microsoft adds ’vibe working’ to Word and Excel; Copilot Agent Mode now default
Business & Startups
From Archives Insiders call AI the 'Ozempic' of music as over half of hip‑hop is AI‑generated
From Archives Altman, Science VP hype AI‑assisted dog cancer vaccine despite lacking proof
From Archives Microsoft cites Fabric IQ to close execution gap for enterprise AI agents
Open Source
From Archives Microsoft names new Copilot chief after AI leadership shakeup
AI Tools & Apps
Common Questions Answered
How do Microsoft's new Critique and Council features enhance Copilot Researcher's capabilities?
Critique and Council transform Copilot Researcher into a multi-model system by introducing a parallel review model that can spot errors and suggest edits. The new features allow the AI to run side-by-side comparisons of outputs, effectively creating a self-checking research assistant that can improve report accuracy and reliability.
What prompted Microsoft to develop the Critique and Council features for Copilot?
The development appears to be a strategic response to the volatile AI market, particularly following the surprise $1 billion Disney deal and OpenAI's abrupt Sora shutdown. Microsoft aimed to create a more robust and flexible AI research tool that can provide multiple perspectives and built-in error checking to address enterprise concerns about AI reliability.
How does the Critique feature specifically work with Copilot Researcher?
The Critique feature integrates Claude as a second model to review outputs generated by the primary Copilot Researcher system powered by OpenAI. This approach allows for a comprehensive review process where one model can identify potential errors, suggest improvements, and provide a cross-validation mechanism for research reports.
Further Reading
- Microsoft rolls out multi-model AI feature 'Critique' along with 'Model Council' tool to expand Copilot capabilities — The Tech Portal
- Microsoft 365 Copilot gets Critique and Council modes — Testing Catalog
- How Microsoft Copilot's 'Critique' feature reduces AI hallucinations — NewsBytes
- Microsoft Copilot Researcher Adds Critique and Council to Improve Trust — Windows Forum
