CTI4AI: Threat Intelligence Generation and Sharing after Red Teaming AI Models
This addresses security vulnerabilities for AI/ML stakeholders like developers and security professionals, but appears incremental as it builds on existing red teaming concepts.
The paper tackles the problem of adversarial attacks on AI/ML systems by developing CTI4AI, a prototype system for methodically identifying and sharing AI-specific vulnerabilities and threat intelligence through red teaming, though no concrete results or numbers are provided.
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.