Microsoft has upgraded its research assistant, Copilot, to use multiple AI models simultaneously. This upgrade significantly improves Copilot's ability to tackle complex research tasks with multiple steps. The new feature, called Critique, combines the power of OpenAI's ChatGPT and Anthropic's Claude to deliver higher-quality results. This is particularly useful for tasks that require factual accuracy, analytical breadth, and presentation, such as those found in academic and professional research settings.
The upgrade scores higher on the Deep Research Accuracy, Completeness, and Objectivity benchmark compared to the most recent Perplexity Deep Research models. This improvement demonstrates the potential of combining multiple AI models to achieve better results. Additionally, Microsoft has introduced the Model Council feature, which allows users to view side-by-side responses from both Anthropic and OpenAI, highlighting areas of agreement and disagreement.
The Critique feature and Model Council option are currently available in Microsoft 365 Copilot's Frontier program, which serves as an early access space for the company's AI innovations. This move highlights Microsoft's commitment to pushing the boundaries of AI research and development.
Microsoft's decision to combine multiple AI models in its research assistant, Copilot, showcases the potential of collaborative AI. This approach can lead to more accurate and comprehensive results, ultimately benefiting research professionals and organizations. The success of this feature may encourage other companies to explore similar collaborations, driving innovation in the AI space.
