Contractual Integrity for Scientific AI-Ready Data Pipelines (CERTIFY) is a three-year, $900,000 grant funded by the NSF’s Cybersecurity Innovation for Cyberinfrastructure (CICI) program. Led by Principal Investigator Charles Cao of the University of Tennessee, Knoxville, alongside NICS co-PIs Tabitha Samuel and Gary Rogers, the project addresses a critical security vulnerability: as scientific data moves across multi-institutional pipelines into AI models, undetected corruption or undocumented changes at any point can compromise downstream discoveries.

To address this challenge, CERTIFY develops a framework that lets datasets carry machine-checkable proof that they meet specified integrity and provenance requirements, so collaborating teams can trust shared data without re-auditing one another’s work. The approach pairs temporal contracts with operating-system-level monitoring and AI agents to detect and contain problems as data moves through multi-institutional pipelines. NICS serves as the core computational engine for CERTIFY, engineering the high-performance computing (HPC) workflows, data management architectures, and automated job scheduling required to operate this integrity framework at scale. Beyond providing foundational HPC infrastructure and domain expertise, NICS shapes the platform’s real-world deployment across complex domains like agricultural sensor networks and electronic health records—ensuring scalable, trustworthy, and fully reproducible data-intensive science.

CERTIFY is supported by NSF award 2613350.