SI² — SuperIntelligence for Science
REFSafE
Evidence-grounded predictive risk analysis and automated safety report generation.
Research question
How can safety reporting remain traceable to operational evidence?
Research approach
Agentic retrieval and generation bring together operational records, historical incidents, and policy evidence. Citation-backed reports and expert feedback support review.
Research outputs & focus
- Evidence-linked safety reports
- Grounded hazard analysis
- Expert-reviewable source attribution
People behind the project
Contributors.
Related publications
The research behind the tool.
REFSafE: A RAG-Enabled Framework for Predictive Risk Analysis and Automated Safety Report Generation in Mission-Critical Environments
Citation
Sanjay Das et al. (2026). REFSafE: A RAG-Enabled Framework for Predictive Risk Analysis and Automated Safety Report Generation in Mission-Critical Environments. RAG4Reports, pp. 47–56.