AI-driven analytics is the use of AI — mainly large language models — inside the analytics workflow: preparing data, writing queries, classifying free text, building reports and explaining results. You ask in plain language; the tool does the routine work and returns an answer you can check.
AI-driven analytics you run on your own infrastructure. Ask Greg, the built-in AI analyst, answers with the SQL it ran; the AI_PROMPT formula classifies thousands of comments in the grid; PII masking keeps personal data away from the model. Works air-gapped.
Traditional business intelligence asks people to translate every question into queries, joins and formulas. AI-driven analytics moves that translation to a model: describe what you want, get back a query, a pivot, a chart or a classification — then check it and keep it.
A typical example is the weekly management report. Without AI an analyst exports data, cleans it, rebuilds the pivots and writes the commentary by hand. With AI-driven analytics the preparation is a saved pipeline, customer comments are categorised by an AI formula, the commentary is drafted for review, and the report refreshes and sends itself on a schedule. The analyst reviews instead of assembling.
The safeguard that matters most is reproducibility: an AI analytics answer should come with the query, formula or pipeline that produced it, so it can be checked, rerun and versioned. ATX AnalyticsTable follows that rule — its AI analyst answers with the queries it ran and proposes objects you place yourself.
Asking "which region's margin fell most last quarter?" and getting back the SQL, the result and a chart; or classifying ten thousand free-text customer comments by topic and sentiment with one formula instead of reading them by hand.
Traditional BI needs someone to build every query, join and report by hand. AI-driven analytics lets people ask in plain language and drafts that work for them — the data model, governance and checking still matter as much as before.
Most BI platforms now ship an AI assistant, including Microsoft Power BI, Tableau, Qlik and Looker. ATX AnalyticsTable by ATX Institute is built for teams that need it on their own infrastructure: it installs locally, works air-gapped and uses the AI model endpoint you choose.
This site needs JavaScript for the full experience. Every page above is reachable without it.