New AI Platform Rescues Lost Research Data, Boosting Open Science

Over 90% of scientific data is lost or unused. Frontiers’ AI-powered FAIR² Data Management system makes research data reusable, verifiable, and citable—helping scientists share responsibly and accelerate global discovery.

For decades, most scientific data has remained locked away — stored on hard drives, hidden in lab notebooks, or lost after projects end. Experts estimate that out of every 100 datasets produced, about 80 never leave the lab, 20 are shared but rarely reused, and fewer than two meet FAIR (Findable, Accessible, Interoperable, Reusable) standards. In the end, only one dataset typically leads to new discoveries.

AI to the Rescue

Now, the open-science publisher Frontiers has launched FAIR² Data Management, the first AI-powered system designed to make every dataset reusable, verifiable, and citable. The platform combines curation, compliance checks, peer review, and interactive data visualization in one seamless process — ensuring that research data is not only preserved but also recognized and reused.

“Ninety percent of science vanishes into the void,” said Dr. Kamila Markram, co-founder and CEO of Frontiers. “With FAIR², no dataset needs to be lost again — every contribution can fuel progress and earn the credit it deserves.”

Smarter, Faster, FAIRer

At the core of FAIR² is an AI Data Steward — powered by Senscience, a Frontiers venture — that performs in minutes what once took months. It organizes, verifies, and formats data automatically, generating four integrated outputs for researchers:

• A certified Data Package
• A peer-reviewed and citable Data Article
• An Interactive Data Portal with visual tools and AI chat
• A FAIR² Certificate ensuring quality and compliance

This approach gives every dataset a permanent home, improves its visibility, and ensures that researchers get the recognition they deserve.

Turning Principles into Practice

FAIR² extends the FAIR principles into an AI-compatible model — ensuring data can be reused by both humans and machines responsibly. It arrives at a pivotal moment, as scientific output grows and artificial intelligence reshapes discovery.

Dr. Sean Hill, CEO of Senscience, said: “Science invests billions in generating data, but most of it is lost — and researchers rarely get credit. FAIR² changes that by making every dataset reusable, recognized, and rewarded.”

Real-World Examples

Frontiers has already launched pilot datasets that show FAIR² in action:

• SARS-CoV-2 Variant Properties: 3,800 spike protein variants linking AlphaFold2 predictions with lab data — aiding pandemic preparedness.
• Preclinical Brain Injury MRI: 343 harmonized MRI scans supporting reproducible neuroscience.
• Environmental Pressure Indicators (1990–2050): A multi-decade dataset enabling climate policy insights across 43 countries.
• Indo-Pacific Atoll Biodiversity: Data from 280 atolls integrating biodiversity, climate, and human-use metrics to guide conservation.

Researchers involved in the pilots said the platform not only preserves data but makes it easier for others to understand and build upon — accelerating progress across disciplines.

Unlocking the Future of Open Science

FAIR² represents a major step toward an open, intelligent, and interconnected scientific ecosystem — one where no discovery is wasted and every dataset adds value.

As Dr. Markram put it, “This is how we unleash science.”