Also known as: Hallucination (AI) · Hallucination (also referred to as confabulation) · Hallucinations
Галлюцинация — сгенерированный ИИ выход, который выглядит правдоподобным или уверенным, но является ложным, неподтвержденным, вводящим в заблуждение или сфабрикованным. В генеративном ИИ галлюцинации могут включать неверные факты, вымышленные источники, ложные цитаты или рассуждения, не основанные на надежных свидетельствах. Их обычно смягчают с помощью grounding, ограничений, проверки, оценки и человеческого контроля.
The production of plausible-seeming but factually incorrect output by a generative AI model that purports to be making an assertion about the real world. For example, a generative AI model that claims that Barack Obama died in 1865 is hallucinating.
A hallucination in generative AI is a confident response by an AI that cannot be grounded by its training data. It may be factually incorrect. In the context of text generation, it's plausible-sounding random falsehoods within its generated text content.
A side-by-side comparison of Hallucination and Confabulation. Understand how both describe plausible but unsupported or fabricated AI output, and why hallucination is often used as the broader operational term.
A side-by-side comparison of Hallucination and Misinformation. It explains how false or fabricated AI outputs differ from false or misleading information shared regardless of intent.
A side-by-side comparison of Unsupported-claim Rate and Hallucination. Understand how a measurement of unsupported claims relates to the broader failure of false or fabricated AI output.