A new ScienceDaily report highlights a publisher‑led push to tackle poor data reuse in science. Frontiers has introduced FAIR² Data Management, a platform with an AI‑powered “data steward” designed to help researchers curate, document and share datasets so they are findable, accessible, interoperable and reusable (FAIR). Frontiers frames the need starkly: out of every 100 datasets, about 80 never leave the lab, 20 are shared but seldom reused, fewer than two meet FAIR standards, and only one leads to new findings.
What’s New
Rather than promising to resurrect data trapped on obsolete media, the FAIR² platform focuses on improving how new and existing datasets are prepared and managed: metadata checks, compliance support, and structured workflows intended to make data reusable and citable across repositories. The ScienceDaily story characterizes the tool as an AI assistant for data curation within publisher workflows.
Why It Matters
Poor data stewardship undermines reproducibility and slows discovery. If tools like FAIR² increase the share of properly curated datasets, downstream reuse and verification could improve—particularly in data‑hungry fields where replication and meta‑analysis are essential. The “90% lost” line should be read as an argument for better data practices, not a literal accounting of irrecoverable research.
How It Works (as described)
The platform provides an AI “data steward” to assist with metadata quality, policy compliance, and curation checks, helping authors and editors ensure datasets meet FAIR standards and are deposited appropriately so they can be discovered and reused. The ScienceDaily write‑up does not detail underlying model architectures nor any capability to scrape or restore non‑digitised archives.
Caveats
- The claims come via a publisher announcement covered by ScienceDaily; they are not a peer‑reviewed evaluation of the platform’s effectiveness.
- The “90%” figure is a rhetorical summary of low reuse/FAIR compliance, not a forensic measurement of permanently lost data.
What’s Next
Impact will depend on adoption by researchers, institutions and repositories, and on independent assessments of whether AI‑assisted curation actually raises FAIR compliance and reuse rates at scale.