# Contested concept: hallucination

_**What you are reading**: one contested-concept page in Deliberus's deliberation graph. Deliberus maps human reasoning — including claims that are contested, mistaken, or extreme — so that disagreement can be examined precisely. A claim's presence here is not an endorsement: the graph records that someone argued it, how it connects to other reasoning, and how far scrutiny has gotten. The metadata on this page describes the state of that scrutiny, not the truth of the claim._

---

- **Contestedness**: semantic
- **Diagnosis**: The term 'hallucination' masks the technical reality that LLMs are always 'predicting' rather than 'retrieving'; the disagreement is about whether the error is a fluke (plausible but wrong) or a fundamental property of the generation process.
- **Canonical**: `/concept/hallucination` (data: `/concept/hallucination.json`)

## Senses in use
- **Plausible-looking but subtly buggy or outdated technical output that requires expert validation.** — used in 3 claim(s)
- **Factually incorrect or nonsensical information presented with high confidence.** — used in 2 claim(s)

## Claims using this term
- [Large Language Models (LLMs) frequently "hallucinate," generating information that is incorrect while maintaining a confident tone.](/claim/claim_3d6cb1434d1d.md) _(sense: Factually incorrect or nonsensical information presented with high confidence.)_
- [Large Language Models (LLMs) frequently generate 'hallucinations,' which are outputs that appear plausible but are factually incorrect or n…](/claim/claim_3f23756cc927.md) _(sense: Factually incorrect or nonsensical information presented with high confidence.)_
- [Computer code generated by Large Language Models (LLMs) often contains subtle bugs that are not immediately obvious to developers.](/claim/claim_69de2bc1bf04.md) _(sense: Plausible-looking but subtly buggy or outdated technical output that requires expert validation.)_
- [Large Language Models (LLMs) generate computer code that appears plausible and functional to a human observer upon initial inspection.](/claim/claim_86971b4c6467.md) _(sense: Plausible-looking but subtly buggy or outdated technical output that requires expert validation.)_
- [Code generated by Large Language Models (LLMs) often appears correct but contains subtle, non-obvious software bugs.](/claim/claim_a692d237d05c.md) _(sense: Plausible-looking but subtly buggy or outdated technical output that requires expert validation.)_