Register Now

    AI Hallucination

    An AI hallucination is an incorrect or imaginative result served up by an AI technology, often reflecting human bias or inaccuracies.

    What Is an AI Hallucination?

    Artificial intelligence (AI) is only as good as the information humans feed it. Therefore, AI technologies sometimes reflect human biases or inaccuracies based on incorrect data inputs. AI hallucinations are often a result of this data quality, as they occur when AI tech serves up information that is made up or incorrect. AI relies on large language models (LLMs) that analyze information to learn over time. Much like the human brain, these machines can learn incorrectly, follow inaccurate data patterns, or come to false conclusions. In these situations, the tech’s outputs might be entirely imaginative. In most cases, organizations that monitor AI technologies quickly address hallucinations when they occur. However, the presence of AI hallucinations is a reminder that humans should always fact check AI outputs for accuracy, consistency, and timeliness.

    See Salsify in Action

    Bring your team, tools, and data together to launch high-impact inbound marketing campaigns in less time.

    Request a demo