Hallucination (AI) — definition and meaning
An AI hallucination is when a model generates information that sounds confident and plausible but is factually wrong or invented. A core limitation of large language models.
Last reviewed 2026-05-25
Hallucination is the term for an AI confidently stating something false — a fabricated fact, a misremembered detail, an invented citation. It happens because LLMs generate statistically likely text, not verified truth. For AI companions, hallucination shows up as a partner 'misremembering' your history or inventing shared events, which is jarring precisely because it breaks the continuity the relationship depends on. It's one of the main things a good memory system mitigates: grounding the model in a verified record reduces how often it invents the past.
About hallucination (ai).
What is hallucination (ai)?
An AI hallucination is when a model generates information that sounds confident and plausible but is factually wrong or invented. A core limitation of large language models.
How is "hallucination (ai)" used in AI companion apps?
Hallucination is the term for an AI confidently stating something false — a fabricated fact, a misremembered detail, an invented citation. It happens because LLMs generate statistically likely text, not verified truth. For AI companions, hallucination shows up as a partner 'misremembering' your history or inventing shared events, which is jarring precisely because it breaks the continuity the relationship depends on. It's one of the main things a good memory system mitigates: grounding the model in a verified record reduces how often it invents the past.
What other terms relate to hallucination (ai)?
Related terms in the AI companion space include: LLM (large language model), Context window, Long-term memory (in AI companions). Each has its own glossary entry on /glossary.
- LLM (large language model)
A large language model (LLM) is an AI system trained on vast amounts of text to predict and generate human-like language. LLMs power AI companions, chatbots, and assistants like ChatGPT, Claude, and Gemini.
- Context window
A context window is the maximum amount of text — measured in tokens — an AI model can consider at once. Anything beyond it is forgotten unless stored and re-supplied by a separate memory system.
- Long-term memory (in AI companions)
An AI companion's ability to recall facts and context from prior conversations across days, weeks, and months.