Large language model (LLM)

A large language model is software trained on enormous amounts of text to predict what words plausibly come next. That prediction is what produces its answers — it is writing the most likely continuation of your prompt, not looking facts up.

It explains both the magic and the danger: an LLM is brilliant at transforming material you give it (summarising, rewriting, extracting) and unreliable when asked to recall facts from memory.

Treating it like a search engine or a database. It does not "know" your numbers — it produces text that sounds like an answer.

Here is the report pasted below. Summarise it in five bullet points for a busy manager, using only what is in the text.