MITRE offers ATLAS (Adversarial Threat Landscape for Artificial-Intelligence Systems), a publicly available knowledge base launched in June 2021 that catalogs the tactics and techniques adversaries use to attack AI and machine learning systems. It is freely accessible to security teams worldwide and is continuously updated.

ATLAS is modeled on the structure of MITRE's ATT&CK framework but covers attack surfaces specific to AI, including data poisoning, prompt injection, model extraction, and adversarial examples. As of November 2025 (version 5.1.0), the framework contains 16 tactics, 84 techniques, 56 sub-techniques, 32 mitigations, and 42 real-world case studies. The framework is built in collaboration with academic researchers, industry partners, and government agencies, and includes free companion tools such as ATLAS Navigator. In October 2024, MITRE's Center for Threat-Informed Defense launched an associated AI Incident Sharing initiative at ai-incidents.mitre.org.