Cloud 101CircleEventsBlog

Publication Peer Review

CSA Large Language Model (LLM) Threats Taxonomy
CSA Large Language Model (LLM) Threats Taxonomy

CSA Large Language Model (LLM) Threats Taxonomy

Open Until: 04/19/2024

This document establishes a common taxonomy and definitions for key terms related to AI risk scenarios, threats, and control measures. The goal is to provide a shared language and conceptual framework to facilitate clear communication and alignment within the CSA AI Control Working Group as objectives and controls are discussed. Additionally, this foundational terminology and taxonomy is intended to assist the CSA AI Risk and Technologies Working Group in its ongoing efforts.

Peer review period has ended.