# Rating Severity for AI-Specific Vulnerabilities

> Learn how to accurately assess and score AI-specific vulnerabilities using hybrid frameworks, moving beyond traditional CVSS to capture behavioral, data, and business impact.

Canonical: https://codelivly.com/rooms/rating-severity-for-ai-specific-vulnerabilities
Last updated: 2026-10-14

| | |
|---|---|
| Level | Advanced |
| Category | Cybersecurity |
| Lessons | 2 |
| Estimated time | 30 min |
| XP | 100 |

This module explores the complexities of scoring AI security vulnerabilities. Traditional CVSS often falls short when evaluating probabilistic threats like prompt injection, data poisoning, or agentic hijacking. You will learn how to apply hybrid scoring models, map findings to the OWASP Top 10 for LLMs and NIST AI RMF, and calculate risk based on blast radius, reproducibility, and tangible business impact.

## Lessons

1. Why Traditional CVSS Doesn't Map Cleanly to AI Risks (30 min)
2. Practical Severity Rating Approaches for AI Findings (30 min)

## Related

- [Learning path: AI Red Team Reporting and Responsible Disclosure](https://codelivly.com/learning-paths/ai-red-team-reporting-responsible-disclosure)
- [All rooms](https://codelivly.com/rooms)