What happens when a question does not map cleanly onto the schema?
It asks. When a question has no clean mapping, dataface.ai shows the candidate columns it can actually see and lets a human choose, rather than guessing and returning a confident wrong number.
It stops and asks, using options that are grounded in your real schema.
The failure mode this exists to prevent
The dangerous output of a text-to-SQL system is not an error. It is a confident number that quietly answered a slightly different question. Nobody catches that in a board deck.
What the user sees instead
An inline card with the candidates. For example: "denied coverage" has no matching column, did you mean denial_category, denial_reason_text, or a prior authorization outcome. The options offered are columns that exist, not invented ones.
The same mechanism covers the other ambiguities
- Which datasource, when more than one plausibly matches
- Credential prompts, when a source needs a session password
- Semantic intent, when the phrasing supports more than one reading
Why a CDO should care about the interruption
The clarification is the audit trail. Someone chose the definition, the choice is attached to the answer, and the next person can see which one was used. That is the difference between a chat toy and something you let near a regulated number.
Where this comes from
See it against your own data
A pilot is scoped to one governed use case and time-boxed to eight to twelve weeks, with success criteria agreed before it starts.
Get started freeRelated questions
- What is dataface.ai?dataface.ai is on-source conversational analytics. Ask a question in plain English, and it generates read-only SQL from your schema metadata and runs it on the systems you already own. No warehouse copy, no pipeline.
- How do you run analytics without moving our data?dataface.ai reads your schema and a few example values per column, generates read-only SQL, and executes it on the source system itself. Your tables are never copied: no pipeline, no warehouse copy, no second place to secure.
- What does the LLM actually see from our data?By default the model sees your catalog: table names, column names, types, and a few example values per column. It never receives your tables, and it only sees query results if a user turns on AI analysis.
- How do you stop an AI from writing to or damaging our database?Generated SQL is SELECT-only. INSERT, UPDATE, DELETE and DDL are blocked, queries are parameterized against injection, and every request passes both a workspace gate and a datasource gate.