Large language model
An AI trained on huge amounts of text that can read and write in plain language. ChatGPT is one. In healthcare they draft notes, summarize, and answer questions, but can be confidently wrong on facts.
A nurse asks a general-purpose chat assistant on the hospital intranet for the steps to adjust a heparin infusion, and gets a clear, well-formatted answer with specific rate changes. The steps come from patterns in the model's training text and differ from the hospital's own nomogram, and nothing in the answer signals that. Before deploying an assistant for protocol questions, require it to answer from the hospital's current documents, show which document each answer came from, and say so when it cannot find one.
Terms like this come up in real clinical scenarios across the HelloAI courses: bite-sized modules with verifiable certificates. An account takes one minute, no password needed.
Sign in →