MITRE ATLAS’ Post

Take a threat-informed approach to secure AI. The Secure AI project accelerates the development of MITRE ATLAS and launches a new AI incident sharing program to boost community knowledge of threats to Artificial Intelligence-enabled systems. ⭐ Expands the database of adversary TTPs and case studies for AI-enabled systems through incident sharing metrics and mechanisms. ⭐ Documented new case studies within ATLAS that address vulnerabilities in industry-relevant systems, including generative AI. ⭐ Describes new relevant mitigations based on documented AI incidents. ⭐ Aligns ATLAS TTPs with the current version of MITRE ATT&CK. We extend our deepest gratitude to our industry partners for their invaluable contributions: AttackIQ, BlueRock.io, Booz Allen Hamilton, Cato Networks, Citi, CrowdStrike, FS-ISAC, Fujitsu, HCA Healthcare, HiddenLayer, Intel Corporation, JPMorganChase, Microsoft, Standard Chartered, and Verizon Business. We invite additional industry participants to join us in this critical endeavor. Your expertise and data contributions will further enhance our understanding of AI threats and bolster defenses across various sectors. Let’s change the game on the adversary! https://hubs.ly/Q02RJrpz0 #threatinformeddefense #cybersecurity #MITREATLAS #AI #AISecurity