Vast amounts of valuable research data remain unused, trapped in labs or lost to time. Frontiers aims to change that with FAIR² Data Management, a groundbreaking AI-driven system that makes datasets reusable, verifiable, and citable. By uniting curation, compliance, peer review, and interactive visualization in one platform, FAIR² empowers scientists to share their work responsibly and gain recognition. Read More
BC
October 17, 2025The “90% of science vanishes” stat is eye-opening and aligns with what I observe in my own testing. I perform experiments across multiple systems daily, creating extensive data about model performance, hardware bottlenecks, and optimization strategies. Most of this data remains in spreadsheets on my local drives, may be mentioned in a forum post, then essentially disappears. Not because it’s useless—I simply lack the time to clean it up and make it usable for others.
The AI automation part sounds promising, especially transforming months of manual work into minutes. But here’s what concerns me: “quality controls and clear summaries that make data easier to understand for general users.” In my experience, AI-generated summaries often miss crucial nuances. When I document why a specific model failed on certain hardware, there are usually odd edge cases and compatibility issues that an AI might gloss over as “performance limitations.” Those specific details are often what other researchers genuinely need.