Glossary
What is large language model (LLM)?
A large language model (LLM) is an AI model trained on large amounts of text, which enables it to understand and generate language. The models behind ChatGPT are an example; they are strong at language, but not at exact calculation.
Large language model (LLM) explained
An LLM repeatedly predicts the most likely next piece of text, a token, based on everything that came before. That lets it write, summarise and translate text and generate code or SQL. It can also produce convincing-sounding nonsense, which is called a hallucination.
Because an LLM predicts text rather than calculating, it is not a reliable calculator and it does not know your business data. For questions about figures it is better to have the model write a query or formula that a database runs, rather than letting the model make up the answer.
For business use, consider where the model runs, what happens to your input and whether it is used to train the model. Business APIs from providers such as OpenAI, Anthropic and Google have different terms for this than the free consumer apps.
Example
An employee at Acme Groothandel asks an LLM for last quarter's revenue. Without access to the data, the model invents a number or declines; with text-to-SQL it writes a query instead, and the database returns the real amount.
Large language model (LLM) in ENABLE
ENABLE uses a language model through the OpenAI API for the SQL Explorer and for "Ask AI" in the Dashboard Builder (pilot). The model receives only metadata and the question, never rows of data or passwords. ENABLE does not offer its own or EU-hosted AI models.
FAQ
Frequently asked questions
What is the difference between an LLM and ChatGPT?
An LLM is the underlying AI model; ChatGPT is an OpenAI app that uses such models. Other well-known LLMs include Claude, Gemini and Llama.
Why can't an LLM calculate reliably?
An LLM predicts text based on patterns and does not perform a real calculation. For exact figures, have the model write a query or formula that a database or calculation tool runs.
Is my input used to train an LLM?
It depends on the provider and the type of account. With business APIs, training on your input is usually excluded by default, but always read the terms.
Further reading
- SolutionAI on your own data: ask questions of business dataQuestion your data warehouse in plain language. The AI sees the structure, not your rows.Read more
- SolutionSecure AI for business: private AI on your dataWhat private, secure AI really means, and how ENABLE limits what the AI sees of your data.Read more
- ResourcesText-to-SQL explained: AI on your databaseText-to-SQL is a technique in which an AI model converts a question in plain language into an SQL query, which is then run on a database. The language model writes the query; the database supplies the figures. How well that works depends mainly on the quality of your metadata and on the safeguards around execution, such as read-only access, timeouts and row limits.Read more
- SolutionAI platform for business: data, AI and appsDashboards, AI on data, custom dashboards and apps under one login and one access model.Read more
Use AI on your own data. Securely, and without the headache.
In an online demo we show how ENABLE makes your data securely available and how you build on it: describing an app in the AI Builder, plain-language questions, dashboards and APIs. And how we set it up and manage it for you.