Skip to content

Concepts

Every concept entity in the Dhabi AI wiki. 4 entries, each with a definition that stands on its own and citations for every number.

  • Answer integrity

    Answer integrity is whether what an AI assistant says about a subject is true and whether it was reached without manipulation.

  • Prompt injection

    Prompt injection is hiding instructions inside content so that an AI system reading that content follows them.

  • Data poisoning

    Data poisoning is inserting crafted documents into the material a model trains on so as to change its later behaviour.

  • Hallucination

    A hallucination is a confident, fluent statement from an AI model that is not true, delivered with the same certainty as a correct one.