AI engineering glossary · Evals guardrails AI security
What is data exfiltration?
Data exfiltration (also called data leakage attack): The unauthorised extraction of confidential information from a system. In language-model apps it often happens through injected instructions that make the model send private data to an attacker, for example by encoding it in a URL or an image link that the interface loads automatically.
Data exfiltration: an example
An injected instruction makes an assistant render a markdown image whose URL contains the user's private notes, which the browser then sends to the attacker's server.
Related terms
More Evals guardrails AI security terms
- Data masking
- Data poisoning
- DeepEval
- Defence in depth
- Denial of wallet
- Direct prompt injection
- Egress filtering
- Elo rating
Where you learn it
Data exfiltration is part of the AI Engineering Program, which teaches it with notes, a hands-on lab and an exam. The first lesson is free.