Log in

LLM (large language model)

An LLM, or large language model, is an AI program trained on huge amounts of text that understands questions and writes answers. ChatGPT, Gemini, Perplexity and Grok all run on LLMs.

How an LLM answers guest questions

The model predicts the most likely answer based on what it learned from text. Many assistants also search the web at the time of the question and link to their sources.

That is why the same assistant may answer slightly differently each time, and answers change when new information appears online.

Hallucinations: when AI gets it wrong

An LLM can state something untrue with full confidence, such as old prices, a pool that does not exist or wrong check-in times. This happens most often when there is little or conflicting information about the property online.

What it means for a hotel

The more consistent, up-to-date facts about your property there are online, the better the chance the model describes it correctly and recommends it to guests whose needs match your offer.

Related terms

See how AI sees your property

The free audit shows whether AI assistants and Google recommend your property.

Run a free audit
Back to the glossary
HotelSpot