Hallucination
Also called: confabulation
Fluent, confident output that is simply false — the defining failure mode of language models.
A hallucination is content the model generates because it is statistically plausible, not because it is true. It is a structural consequence of next-token prediction rather than a bug to be patched: the model optimises for plausible, and plausible and true usually coincide, right up until they do not. Grounding, retrieval, and verification reduce it; nothing eliminates it.
In practice: Invented case citations, invented DOIs, invented API endpoints — all classic, all confidently formatted.
Where this comes up
- AI Bible Study: How to Use AI as a Careful Study Aid
- AI Book Recommendations: How to Get Picks That Truly Fit You
- AI Career Path for Beginners: Your Comprehensive Guide
- AI Consultant Certification in 2026: Skills, Training, and Career Path
- AI Crochet Patterns: What You Need to Know Before You Stitch
- AI Data Analytics Course in 2026: What Analysts Should Learn First and How to Choose a Course