Murata unveils AI-driven R&D platform, claims 22.5× speedup

Murata's materials development work generates varied experimental data — material compositions, process conditions, evaluation results, and analysis values — often managed in different formats depending on the theme or site. That fragmentation meant researchers spent time locating and reusing data that already existed inside the company. The new platform is Murata's attempt to break those silos by connecting its existing data bases and knowledge base to multiple AI agents, so that engineering knowledge accumulated over time can be reused beyond team and organizational boundaries.

The workflow the company describes has researchers specify the purpose and data needed on a browser screen. AI agents then take over a chain of tasks: referencing and analyzing experimental data, generating and running analysis code, visualizing data, drafting considerations, and writing reports. Murata says the final judgment stays with the researcher, who confirms the content — a division of labour where the agents handle the legwork and the human reviews the output.

How much of this becomes routine practice likely hinges on whether the agents can reliably handle data across different formats and sites, and on how researchers trust and adopt the results. Murata's own tests offer some signals — an about-10-minute run through data reading, chart creation, and report writing, and the 22.5× estimate for the full planning-to-report workflow — but those are internal measurements under a simulated scenario, so the real question is whether the same gains hold in day-to-day projects.