A case for how to shape ‘ingredient lists’ for AI models

A policy paper from the Institute for Security and Technology recommends developing AI bills of materials (AIBOMs) to enhance transparency and reduce cyber risk, arguing that foundational work and common standards are needed before widespread adoption. The paper outlines supply- and demand-side approaches, warns against fragmented implementations, and positions AIBOMs as a starting point for policy discussions rather than a complete solution to AI security challenges.