How AI interviews work
Learn how AI interview questions run in a conversational survey.
This article explains how AI interviews work. It covers adaptive follow-up questions, prompt setup, context across multiple turns, and how the AI moderates in different languages. It also covers privacy and moderation considerations.
Before you start
Create or open a survey in the editor first. Decide whether you want a text or voice interview before you configure the model and length.
Steps
Add an AI interview question
In the survey editor, click + Add. Choose AI interview. In the two-step setup, enter what you want to learn. Then choose Text chat or Voice under How should it be conducted?.English interface · Demo study. Use the controls described in this step.
Configure the AI interviewer
Click Interviewer. Choose a personality or create your own with a name, description, and example questions that guide the style. To read the prompt behind it, open AI model > Advanced prompt > Configure.Set the interview length mode
Under Length, choose how the AI decides when to stop follow-up questions. Use Fixed Count to set an exact number of follow-up questions. Use AI Decides to let the AI stop when it has enough depth, guided by min/max limits and your stopping criteria. Use Time-Based to run the interview for a set duration. Voice interviews are time-based only.Choose a model
Select the AI model for the interview from the model picker. Existing questions keep their saved selection. Text and voice have separate pickers. If needed, run a small pilot and compare results. See Choosing a model for trade-offs.Pick the response quality
Under AI model, set Response quality for how text follow-up questions are generated. Fast (default) generates a single question in one step. It is the most responsive option. Slow generates five candidate follow-up questions, grades each with personality-specific criteria, and picks the highest-ranked one. Use it when question quality matters more than latency.Choose what it can refer to (optional)
Click What it can refer to to let the AI use the respondent's earlier answers, or insights from this survey or other surveys, during the interview. The AI can use these references for more contextual follow-up questions. You can also add per-respondent context that feeds into the interview dynamically.Monitor and review
After launch, review AI interview transcripts in Results. Check the full exchange between the AI and each respondent.
Details and limitations
AI interviews adapt follow-up questions based on participant answers. The system probes for clarity and depth. This can produce richer qualitative data than a standard open-text field.
The AI preserves context across conversation turns. Follow-up questions can refer to what the respondent already said. This creates a more natural interview flow.
Three question modes control the length of a text interview: Fixed Count (exact number of follow-ups), AI Decides (AI stops when it has enough depth, with configurable min/max and stopping guidance), and Time-Based (runs for a set duration). Voice interviews always run for a set duration.
Response quality controls how text follow-up questions are produced. Fast (default) generates a single question in one step and works best when responsiveness matters. Slow generates five candidate questions, grades each with personality-specific criteria, and picks the highest-ranked one. Use it for important interviews where question quality matters most.
Choose the respondent language in the survey editor. Text and voice have different language and provider support; check the current voice settings and test your selected language with the chosen transcription, conversation, and speech providers before recruitment.
What it can refer to lets the AI use answers from earlier questions in the survey. This helps follow-up questions build on what the respondent already shared in previous sections. Per-respondent context can also be injected dynamically.
You can read the full prompt the interviewer uses, edit its context, add your own sections, and review every exchange in the response data.
Check the result
You should have an AI interview question configured with an interviewer, a length mode, and a model selection. After launch, you should be able to review full transcripts in Results.
If something goes wrong
If the interview does not behave as expected, reopen the question and check the selected mode under Length. Voice interviews only support Time-Based.
