Definition
An LLM predicts useful continuations of language based on patterns it learned during training. It can be prompted with instructions and context, but it does not automatically know your current business information.
Why it matters
LLMs power many modern AI experiences, yet the surrounding system determines whether the output is useful and safe.
Business example
An LLM helps a support representative turn a technical resolution into a customer-ready explanation.
When to use it
Use it when the input or output is primarily language and some variability is acceptable.
When not to use it
Do not use it as a source of record for numbers, policy, or private knowledge without retrieval and validation.
How Automathing approaches it
We treat the LLM as one component: its prompt, context, tools, evaluations, and user experience all matter.