Introduction to AIVEX
Remediation priority, traditionally focused on Software Bill of Materials (SBOMs) and Vulnerability Exploitability eXchange (VEX) statements, is no longer sufficient in today's environment. The rise of AI and autonomous robots has created a need for a more comprehensive approach to vulnerability triaging.
AIVEX, a new triage model, aims to address this issue by providing context to vulnerability remediation priority. Developed by independent researcher and security architect Devashri Datta, AIVEX combines two new elements: a safety relevance interpretation layer (SRIL) and an extension to the CycloneDX VEX schema.
Limitations of Traditional SBOM, VEX, and CVSS Scores
Traditional SBOM, VEX, and CVSS scores have limitations in providing context to vulnerability remediation priority. SBOMs list the components within the software, while VEX statements indicate whether known vulnerabilities are exploitable. CVSS scores, used as a severity indicator, do not provide sufficient context for AI-driven systems.
The lack of context reduces the effectiveness of remediation priority, and the expansion of AI software will magnify the problem. Supply chain attacks will continue to grow, and a new solution is urgent.
AIVEX and SRIL
AIVEX is a proposed extension to the CycloneDX VEX schema, making SRIL machine-readable in structured fields for model provenance, inference-time attack surface classification, safety domain annotation, and AI lifecycle stage. SRIL provides four dimensions of context: safety domain classification, lifecycle stage mapping, consequence severity modifier, and exploitability in context.
SRIL is a manual effort required from the DevSecOps team, but one that is fully justified by the potential blast radius of an unpatched low-severity AI vulnerability causing robotic third-party harm. Flexera and Anchore have adopted SRIL and are working on shipping it to customers.
Benefits of AIVEX and SRIL
AIVEX and SRIL provide more realistic triaging and benefits for AI regulatory compliance. The US National Institute of Standards and Technology (NIST) Secure Software Development Framework promotes risk-informed decisions, and AIVEX/SRIL operationalize this guidance by clarifying how SBOM and VEX data feed into real-world governance decisions.
The EU AI Act and NIST's AI Risk Management Framework emphasize governance processes that account for operational context and real-world impact of AI system failures. AIVEX/SRIL address this need by providing a structured mechanism to connect vulnerability data to safety consequence.
Conclusion
AIVEX, a new triage model, addresses the limitations of traditional SBOM, VEX, and CVSS scores by providing context to vulnerability remediation priority. By combining SRIL and AIVEX, organizations can generate a safety-adjusted priority, reflecting not just how severe a vulnerability is in isolation but how much it matters in the specific operational context where affected software is deployed.
As AI continues to transform industries, the need for a more comprehensive approach to vulnerability triaging becomes increasingly important. AIVEX and SRIL offer a solution to this problem, providing a more effective way to prioritize vulnerability remediation and reduce supply chain threats.
- Related: Are SBOMs Failing? Supply Chain Attacks Rise as Security Teams Struggle With SBOM Data
- Related: SBOM Pioneer Allan Friedman Joins NetRise to Advance Supply Chain Visibility
- Related: AI and Cybersecurity – Everything You Wanted to Know, But Were Afraid to Ask
- Related: The Wild West of Agentic AI – An Attack Surface CISOs Can’t Afford to Ignore
Source: SecurityWeek