{"head":{"title":"What Is an AI Hallucination? Definition | OceanLab","description":"An AI hallucination is when a model states something false or made up as if it were true. Why it happens, how products reduce it, and common mistakes.","canonical":"https://oceanlab.in/glossary/hallucination"},"crumbs":[{"name":"Home","path":"/"},{"name":"Glossary","path":"/glossary"},{"name":"hallucination","path":"/glossary/hallucination"}],"slug":"hallucination","term":"Hallucination (AI)","aka":["confabulation"],"category":"AI building blocks","short":"When an AI model produces information that is false or made up, presented as if it were true.","body":[["A language model writes the most likely continuation of the text, not a looked-up fact. When it lacks the information, it can still produce fluent, confident text: a citation that does not exist, a wrong figure, a policy that was never written."],["It cannot be removed entirely, but it can be made much rarer and easier to catch. Grounding answers in your own data with ",{"href":"/glossary/rag","text":"RAG"},", using ",{"href":"/glossary/tool-calling","text":"tool calling"}," for live facts, asking for citations and checking them, and allowing the model to say it does not know all help."]],"why":["In business software a confident wrong answer can be worse than no answer: a wrong dosage, a wrong contract term, a wrong refund amount. Products that use AI for facts need a design that makes errors visible and checkable, plus ",{"href":"/glossary/human-in-the-loop","text":"human review"}," where the stakes are high."],"example":["An HR assistant was asked about parental leave and described a policy the company does not have. After the fix, it answers only from the retrieved policy document, quotes the section and says it cannot find an answer when the document is silent."],"inBuild":null,"mistakes":[["Trying to fix it with prompt wording alone instead of grounding and checks."],["Hiding sources from users, so they cannot verify answers."],["Not measuring it. Track the rate in ",{"href":"/glossary/llm-evals","text":"evals"}," and in production samples."]],"related":[{"href":"/glossary/rag","label":"RAG (retrieval-augmented generation)","detail":"A pattern that finds relevant passages from your own data and gives them to a language model along with the question, so it answers from your content."},{"href":"/glossary/guardrails","label":"Guardrails (AI)","detail":"Checks placed around an AI model that block, fix or flag unsafe, off-topic or incorrect inputs and outputs."},{"href":"/glossary/llm-evals","label":"LLM evals (evaluations)","detail":"Repeatable tests that score how well a model, prompt or AI feature does a specific task."},{"href":"/glossary/temperature","label":"Temperature (LLM setting)","detail":"A setting that controls how random a language model's output is: lower is more predictable, higher is more varied."}],"tools":[{"href":"/tools/rag-cost-calculator","label":"RAG cost calculator","detail":"Free tool"}],"links":[{"href":"/ai","label":"AI by industry","detail":"AI use cases, costs and risks per industry."},{"href":"/ai-automation","label":"AI automation","detail":"How OceanLab automates company workflows with AI."}],"sources":[],"legal":false,"updatedAt":"2026-10-25"}