This initiative kicks off with the critical development of a rigorous scoring system specifically for the OWASP Agentic AI Top 10. This initial, high-impact deliverable will then expand into a comprehensive Artificial Intelligence Vulnerability Scoring System (AIVSS) Framework Package. The broader project aims to provide a structured and quantifiable methodology to identify, assess, and mitigate vulnerabilities specific to all types of AI systems – not just Large Language Models (LLMs), Generative AI, or Agentic AI. The ultimate goal is a complete AIVSS Framework Package that serves as a baseline for understanding and managing AI security risks across the entire AI landscape.

Key Deliverables

1. Agentic AI Top 10 Vulnerability Scoring System:
   • A precise and quantifiable scoring methodology tailored to the unique risks identified in the OWASP Agentic AI Top 10.
   • Clear rubrics and guidelines for assessing the severity and exploitability of these specific vulnerabilities.
2. Comprehensive AIVSS Framework Package:
   • Standardized AIVSS Framework: A scalable framework validated across a diverse range of AI applications, including and extending beyond Agentic AI.
   • AIVSS Framework Guide: Detailed documentation explaining the metrics, scoring methodology, and application of the framework.
   • AIVSS Scoring Calculator: An open-source tool to automate and standardize the vulnerability scoring process.
   • AIVSS Assessment Report Templates: Standardized templates for documenting AI vulnerability assessments.

The Importance of this Top-Level OWASP Project: Meeting Critical Demands in AI Security

This Top-Level OWASP project is strategically established to meet the following critical demands in the rapidly evolving field of Artificial Intelligence security:

• Demand for a Focused Agentic AI Scoring System & Comprehensive AI Vulnerability Quantification: There is an immediate and pressing need for a rigorous scoring system for the OWASP Agentic AI Top 10. Beyond this, the broader AI landscape requires a method to move beyond securing specific implementations and quantify the offensive potential and vulnerabilities inherent in underlying AI technologies themselves. This project directly addresses this fundamental gap.
• Demand for Proactive Measures Against Emerging Threats (Future-Proofing): As AI technology evolves and diversifies (e.g., advancements in Agentic AI), there is a demand for a security framework that can adapt to new models, algorithms, and applications. This project provides such a future-proof approach, ensuring OWASP remains at the forefront of addressing emerging AI threats.
• Demand for Open Collaboration and Knowledge Sharing: Effective AI security requires broad collaboration. This open-source OWASP project fulfills the demand for a platform that fosters knowledge sharing among security professionals, AI researchers, and industry stakeholders across all AI domains.
• Demand for Globally Applicable, Technology-Agnostic Solutions: The security of AI systems is a global concern. This project addresses the demand for a technology-agnostic framework that can achieve wide reach and impact, helping to improve the security of AI systems worldwide.
• Demand for Alignment with OWASP’s Core Mission in the AI Era: OWASP’s mission to secure software must extend to AI. This project directly meets the demand for a framework focused on identifying, assessing, and mitigating vulnerabilities in this increasingly critical technology.

Road Map

The following is the initial roadmap:

1. AIVSS Core Definition & Agentic AI Top 10 Scoring (Months 1-3):
   • Define core AIVSS metrics, with an initial focus on metrics directly applicable to scoring the OWASP Agentic AI Top 10. Ensure clarity and precision.
   • Develop initial scoring rubrics for these Agentic AI-focused metrics.
2. AIVSS Framework Specialization & Expansion (Months 4-6):
   • Develop specialized scoring rubrics for other specific AI system types (beyond Agentic AI).
   • Identify factors unique to each AI type that influence vulnerability assessment.
   • Create templates for AIVSS assessment reports, adaptable to different AI system types.
3. AIVSS Scoring Calculator Development (Months 7-9):
   • Develop the core functionality of the AIVSS scoring calculator, ensuring it supports core AIVSS metrics, Agentic AI Top 10 scoring, and specialized rubrics.
   • Implement the ability to add new AIVSS metrics and scoring rubrics.
4. AIVSS Tool Testing and Refinement (Months 10-12):
   • Test the AIVSS scoring calculator against a diverse set of AI systems (including Agentic AI test cases), generating assessment reports.
   • Refine metrics, rubrics, and the calculator based on user feedback and assessment results.
5. Documentation and Release (Month 12):
   • Finalize the AIVSS Framework Guide (comprehensive, clear, easy to understand).
   • Release the AIVSS Scoring Calculator as an open-source project.
   • Publish AIVSS assessment report templates and example reports (including for Agentic AI) on the OWASP project website.
   • Create detailed project documentation, including an integration guide for existing SDLCs.

Multi-Year Project Roadmap

• Year 2: Apply AIVSS to Financial and Healthcare Industries:
  • Develop industry-specific guidelines for applying AIVSS to AI systems in finance and healthcare.
  • Create case studies and assessment reports showcasing AIVSS application in these sectors.
  • Collaborate with industry experts for validation and alignment with best practices.
• Year 2/3: Expand AIVSS for Emerging AI Threats:
  • Continuously update the AIVSS framework to address new AI security threats, including further evolutions in Agentic AI vulnerabilities beyond the initial Top 10.
  • Develop new AIVSS metrics and scoring rubrics as needed.
• Year 3+: AIVSS Certification Program:
  • Explore creating a certification program for professionals proficient in using the AIVSS framework to drive adoption and enhance AI security expertise.