SECAI Core · phase 14 of 15 · noun
AI Vulnerability Database
An open-source knowledge base that collects the data about failures modes for AI models, datasets, and systems
The Explain card
- Plain English
- The AI Vulnerability Database is an open-source knowledge base that collects information about failure modes in AI models, datasets and systems, so practitioners can learn what goes wrong and how.
- Example
- While assessing a vision model, a tester consults the database, finds documented weaknesses affecting that model family, and adds the corresponding evasion tests to the assessment plan.
- Why it matters
- Traditional vulnerability feeds do not cover problems like poisoned datasets or models that fail on certain inputs. A dedicated database fills that gap for AI defenders and red teams.
- Hook
- A CVE list for things that were never just code.
Word knowledge
How the word is built, where it came from, and what it sits beside in memory.
In a sentence
The red team consulted the AI Vulnerability Database before testing a vision model for adversarial-image weaknesses.
Why these words
- AI English initialism of artificial intelligence identifies the technological field that modifies vulnerability database
- vulnerability Late Latin vulnerabilis, capable of being wounded specifies the type of entries associated with the database
- database English compound of data and base serves as the head noun and names the organized resource
Where it came from
- Origin
- English technical phrase formed from AI, vulnerability, and database, and used as the expanded name behind AVID
- Entered the language
- c. 2022
- What changed
- AI abbreviated artificial intelligence, vulnerability broadened from physical susceptibility to technical weakness, and database became a closed compound for an organized digital collection.
How it is spelled
- Pattern
- AI remains uppercase because it is an initialism
- Pattern
- database is written as one closed compound
- Pattern
- vulnerability ends in the noun-forming suffix -ity
Spelled like
- API
- URL
- website
- filesystem
- reliability
- availability
Broken into chunks
-
AI
- AI model
- AI safety
-
vulner
- vulnerable
- invulnerable
-
ability
- reliability
- scalability
-
data
- dataset
- metadata
-
base
- codebase
- knowledge base
What it sits beside
Same subject
- adversarial example
- model robustness
- prompt injection
- red teaming
Same shape
- AI Incident Database
- National Vulnerability Database
- Common Vulnerabilities and Exposures
security knowledge resources
- AI Vulnerability Database
- National Vulnerability Database
AI security terms
- prompt injection
- model extraction
- data poisoning
Where it sits in the deck
Phase 14: Incident Response, Evaluation, and Knowledge Bases
When detection fires, teams need structured response plans, scoring frameworks, and curated knowledge bases to triage, measure, and learn from AI security events.
