CTI4AI: Threat Intelligence Generation and Sharing after Red Teaming AI Models

TL;DR

This paper introduces CTI4AI, a prototype system designed to identify, share, and enhance AI/ML security threat intelligence, addressing vulnerabilities and fostering collaboration among stakeholders.

Contribution

The paper presents a novel prototype system for systematic AI/ML threat intelligence generation and sharing, improving security practices in AI ecosystems.

Findings

Prototype demonstrates effective threat intelligence sharing

Enhances collaboration among AI security stakeholders

Identifies key vulnerabilities in AI/ML models

Abstract

As the practicality of Artificial Intelligence (AI) and Machine Learning (ML) based techniques grow, there is an ever increasing threat of adversarial attacks. There is a need to red team this ecosystem to identify system vulnerabilities, potential threats, characterize properties that will enhance system robustness, and encourage the creation of effective defenses. A secondary need is to share this AI security threat intelligence between different stakeholders like, model developers, users, and AI/ML security professionals. In this paper, we create and describe a prototype system CTI4AI, to overcome the need to methodically identify and share AI/ML specific vulnerabilities and threat intelligence.