Overview of the Initiative
MITRE’s Center for Threat-Informed Defense has launched the AI Incident Sharing initiative. This program brings together over 15 companies to enhance understanding of threats and defenses related to AI systems. It is part of MITRE’s Secure AI project, which focuses on secure collaboration regarding AI-related threats, attacks, and accidents. The initiative also aims to expand the MITRE ATLAS knowledge base that has been gathering anonymized incident data for two years. Participants will receive protected data on real-world AI incidents, allowing for better risk analysis.
Key Features
- Anyone can submit AI incident reports through a dedicated web portal.
- Submitting organizations may gain membership to enhance data-driven intelligence.
- The ATLAS framework has been updated to include generative AI threat information.
- New case studies and attack techniques related to generative AI have been added.
Importance of the Initiative
This initiative is critical for improving collective defense against AI-related threats. By sharing standardized information quickly, organizations can better protect themselves and mitigate risks. The collaboration spans multiple industries, including finance, technology, and healthcare, underscoring the widespread impact of AI threats. The initiative mirrors MITRE’s successful aviation safety program, demonstrating its potential to create safer environments across various sectors.
