Скачать или смотреть AI 101 in Healthcare | LLM Hallucinations: Causes, Risks & Control | Session 1.40

Large Language Models (LLMs) such as ChatGPT, Bard, and Bing Chat are powerful—but they can also hallucinate, generating confident yet incorrect information. In healthcare, such errors can have serious consequences.

In Session 1.40 of AI 101 in Healthcare, we explain what LLM hallucinations are, why they occur, and how they can be systematically reduced in real-world AI applications.

🔍 What you’ll learn in this session:

What AI hallucinations are and how they differ from simple errors
Real-world examples of LLM hallucinations (science, aviation, and AI history)
Types of hallucinations: factual, prompt-based, logical, and contextual
Why hallucinations happen: data quality, generation methods, and context gaps
How prompt engineering, temperature control, and multi-shot prompting reduce hallucinations
Why hallucinations are a critical risk in healthcare AI systems

This session is essential for clinicians, data scientists, AI developers, researchers, and healthcare leaders aiming to build safe, reliable, and trustworthy AI systems.

📌 Part of the AI 101 in Healthcare – Season 1 series