OWASP co-leads release free RAID-AI ransomware assessment
Original title: New Framework Puts Ransomware Resilience for AI Systems in Focus
Users manually rate controls from zero to five; RAID-AI does not connect to systems or verify the answers.
Behnaz Karimi and Yuvaraj Govindarajulu, co-leads of OWASP's AI Bill of Materials initiative, released the Ransomware Assessment and Incident Defense Framework for AI Systems (RAID-AI) free online after about two years of development. The web console asks CISOs and engineers to score general and AI-specific controls themselves on a zero-to-five scale and maps results to NIST CSF 2.0, CISA CPG 2.0, MITRE ATT&CK, MITRE ATLAS, and NIST Adversarial guidance. Organizations do not connect systems or upload documents. Karimi said she hopes later to add implementation resources and a similar tool for agentic AI.
Unconfirmed: Karimi's comments that ransomware is hitting AI systems, including an alleged June 2026 Novo Nordisk incident, are her citations rather than confirmed findings in this report.
- A planned ISC2 Security Congress debut in Denver on October 26-28 did not happen because of visa issues.
- Controls cover recovery, training, sensitive data, and AI data, tools, and runtime.
- Dashboards and downloadable PDF reports show maturity and framework mappings.
- Results depend on self-assessment and are not an automated audit.
- Products
- RAID-AI
From the original article
2 min read at infosecurity-magazine.com
RAID-AI aims to help organizations assess their security preparation for ransomware attacks targeting AI systems
By Kevin Poireault · infosecurity-magazine.com
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.