Executive Summary
An arXiv study on AI supply-chain security highlights risks from reused pretrained models and public datasets whose provenance and prior compromises developers may not understand.
What Happened
This research examines AI supply-chain security and emphasizes risks from reused pretrained models and public datasets. Developers may inherit weaknesses without fully understanding component provenance, quality, embedded threats, or prior compromises.
Vulnerability / Attack Technique
AI supply-chain risk (research)
Technical Mechanism
Research on component provenance.
AI Security Category
LLM03:2025 Supply Chain
Part of the AI Supply Chain and AI SBOM topic.
Relevant Frameworks
RealGround Analysis
Provenance is often unknown. RealGround analysis: record model and dataset provenance in an AI SBOM.
Recommended Controls
- Track model and dataset provenance
Related RealGround Service
AI Agent SBOM and LLM Mapping · or start with the free AI Agent Security Scanner.
Sources
arXiv: https://arxiv.org/html/2512.23385v1
Summary and analysis written by RealGround from the source above; last updated 2026-10-18.