Microsoft Who evaluates the evaluators? The data science behind agent evals Microsoft Who evaluates the evaluators? The data science behind agent evals At Microsoft Copilot Studio, we talk a lot about ways to make agents better. You can add knowledge sources, fine-tune models, insert prompts, and more. But how do you know whether an agent is… Read More »
Frontier Tuning: Teaching AI to work the way you do Frontier Tuning: Teaching AI to work the way you do Today at Microsoft Build, we introduced Frontier Tuning, a new approach to making AI work the way your business does by applying reinforcement learning inside your compliance boundary with your own data, processes, and conventions. We’re announcing private preview, available through Forward Deployed Engineers, and upcoming availability in Microsoft… Read More »
Mistral joins Copilot Studio’s growing lineup of model providers Mistral joins Copilot Studio’s growing lineup of model providers As organizations around the world continue to bring more agents into production, there’s a growing demand for AI models that align with regional expectations around data handling and control. Microsoft Copilot Studio aims to meet this need by combining… Read More »
New and improved: Computer-using agents, a new workflows experience, and real-time voice experiences New and improved: Computer-using agents, a new workflows experience, and real-time voice experiences Expectations for agents are changing quickly. Teams want to move beyond conversational experiences to systems that can help get work done by interacting with applications, executing workflows, collaborating across… Read More »