What is 'hallucination' in the context of large language models like Claude?
Select an answer to reveal the explanation.
Short Explanation
Hallucination is when Claude sounds confident but gives you wrong or made-up facts. It’s like a student who doesn’t know the answer but writes something convincing anyway.
Full Explanation
Hallucination refers to a model generating information that is factually incorrect, fabricated, or unsupported by its training data, but presented with apparent confidence. This can manifest as fake citations, incorrect statistics, invented historical events, or plausible-sounding but wrong technical details. The option describing Claude refusing to answer due to safety filters is not hallucination; it is an intentional safety behavior, the opposite of generating fabricated content. The option describing incorrect image processing is a vision processing error, which is a different type of error. The option describing an API error code due to rate limiting is an infrastructure issue, completely unrelated to the model's output quality.