AI engineering glossary · Evals guardrails AI security
What is vector and embedding weaknesses?
Vector and embedding weaknesses: An OWASP-listed risk area covering flaws in how embeddings and vector databases are used in retrieval systems, such as weak access control across users, poisoned documents in the index and attacks that reveal source text from embeddings.
Vector and embedding weaknesses: an example
In a shared vector store, a query from one client retrieves another client's confidential chunks because filtering by tenant was missing.
Related terms
More Evals guardrails AI security terms
- Verbosity bias
- Access control
- Adversarial example
- Adversarial robustness
- Adversarial suffix
- Adversarial test set
- Adversarial training
- Answer relevancy
Where you learn it
Vector and embedding weaknesses is part of the AI Engineering Program, which teaches it with notes, a hands-on lab and an exam. The first lesson is free.