Embedded Knowledge Exposure

Type: technique

Description: The adversary intentionally interrogates an AI system about its embedded knowledge. Since some AI systems fully share these resources when explaining certain answers, full knowledge sources or artifacts can sometimes be listed and even leaked if targeted directly by an adversary.

Version: 0.1.0

Created At: 2024-10-03 22:24:49 +0300

Last Modified At: 2024-10-03 22:24:49 +0300

External References

  • --> Discovery (tactic): An adversary can discover information that's been embedded in the AI system under the misconception that it would only be used for training and wouldn't be directly accessible to the AI system's users.