- Multi-model agents will verify each other before sharing their research with you to ensure maximum quality
- Researcher mode with Critical enabled scores high on the DRACO benchmark
- Copilot Cowork is here for Frontier Program customers
Microsoft has announced plans to upgrade its M365 Copilot Researcher agent with a clear focus on using multiple models in AI workflows to combine the power of various systems.
As part of this move away from single-model systems, multiple AI agents will collaborate and assign different parts of the task to each other.
To begin, the research agent will use GPT models to generate the initial response, with Claude stepping in to review it for accuracy, completeness, and quality.
Article continues below
The M365 Copilot finder agent will transmit responses via other agents
Jared Spataro, marketing director of Microsoft AI at Work, explained that the update follows the success of Anthropic’s Claude Cowork, which has since been integrated into M365 Copilot. The aptly named Copilot Cowork, now available in the Frontier program in anticipation of wider deployment, allows humans to delegate work to AI.
Spataro explained how Copilot Cowork moves the utility of AI from simple, basic prompts to end-to-end task execution, ideal for long-running, multi-step workflows.
As for Researcher mode with the new Claude-based Critic feature, it has already outperformed single-model systems in early tests, with the latter ensuring higher quality output. It scores 13.8% higher than the DRACO (Deep Research Accuracy, Completeness and Objectivity) benchmark, considered the industry standard.
Achieving 57.4% using the multi-model configuration, it is more than twice as reliable as Deep Research with OpenAI’s o4-mini model. It’s also better than o3-based Deep Research, Gemini Deep Research, Claude Opus 4.6, and Perplexity’s Deep Research when using Opus 4.5 and 4.6. Microsoft did not compare it to newer flagship models like GPT-5.4, acting in a singular way.
“When intelligence and trust go hand in hand, AI stops being an experience and starts becoming a work tool,” Spataro wrote, speaking about Microsoft’s progress toward the third wave of M365 Copilot – an intelligence he defines as “understanding.”[ing] the context of the work.”
Follow TechRadar on Google News And add us as your favorite source to get our news, reviews and expert opinions in your feeds. Make sure to click the Follow button!
And of course you can too follow TechRadar on TikTok for news, reviews, unboxings in video form and receive regular updates from us on WhatsApp Also.
