Glossary term
Hallucination
Output from a generative model that is fluent and plausible but not supported by its inputs or by fact.
Language models generate likely text, not verified truth. Without grounding they may invent details, citations or figures.
Mitigations include retrieval with citations, abstention rules, constrained outputs, validation and human review for consequential uses.
Related terms
- GroundingConstraining a model’s output to information in supplied sources, so that claims can be traced and verified.
- GuardrailsRuntime checks on the inputs and outputs of an AI system that block, modify or escalate content that breaks defined rules.
- Evaluation setA curated collection of realistic inputs with definitions of correct output, used to measure AI system quality.
Turning concepts into a working system?
We help teams choose the right pattern and build it with evaluation from day one.