An interview with a real person where an AI, not a human researcher, asks the questions, follows up on the answers and records the conversation. The respondent is real and only the moderator is automated. That makes it a different method from synthetic research, where the person answering is AI-generated too.
You write a discussion guide and bring the participants, either your own customers or people from the platform’s panel. Each participant joins when it suits them, by voice, video or text. The AI moderator works through the guide, asks a follow-up when an answer is thin, and records the session. Afterwards the platform transcribes each interview and usually summarizes the themes across them.
The appeal is scale. A human moderator can run a handful of interviews a day. An AI moderator can run many at once, in several languages, at any hour. What stays the same is the recruiting: every interview still needs a real person to show up.
The two sound alike because both have an AI asking the questions. The difference is who answers.
| AI-moderated interview | Candor auto-interview | |
|---|---|---|
| Who asks the questions | An AI moderator | An AI interviewer |
| Who answers | A real, recruited person | A synthetic participant built from evidence |
| What the answers are | Real data about that person | An estimate of how a type of customer reasons |
| What you need first | Participants to recruit or a panel to draw on | A description of your audience |
| What it is judged on | Moderator quality, follow-up probing, transcription | What participants are built from, and whether they hold their views |
In a Candor auto-interview, an AI interviewer works through a research plan with each synthetic participant, follows up on what they say, and moves on once each learning goal is covered. When the interviews finish, Candor writes a report across all of them. Nobody is recruited, because nobody real is interviewed.
Use AI-moderated interviews when you need to hear from real customers and want more conversations than your team can moderate. The answers are real, so they can support a decision on their own.
Use synthetic participants earlier, when you are still narrowing down: many concepts, many questions, and no time or budget to recruit for each. They show you where to look. The two work well in sequence, with a synthetic round deciding which questions deserve real recruiting. The guide to synthetic research tools draws this boundary in its section on what is not synthetic research, and where synthetic research fits in your workflow shows how teams sequence the two.
For the synthetic side of the line, read what synthetic user research is and how Candor works. To compare Candor with a platform built around real participants, see Candor vs UserTesting.
You write a discussion guide and recruit participants, or use the platform's panel. Each participant joins by voice, video or text at a time that suits them, and an AI moderator asks the questions, follows up when an answer is vague, and moves through the guide. The platform records and transcribes each session and usually summarizes themes across them. Many interviews can run at once, which is the main advantage over scheduling a human moderator for each one.
No. In an AI-moderated interview the person answering is real, so the answers are real data about that person. In synthetic research the person answering is AI-generated, so the answers are estimates of how a type of person would respond. The two are often listed side by side, but they answer different questions and should be judged on different criteria.
No. Candor's auto-interviews use an AI interviewer, but every participant is synthetic. Candor does not recruit or interview real people. If you need to hear from real customers at scale, an AI-moderated interview platform is the tool for that job, and many teams use Candor first to decide which questions are worth the recruiting cost.
Use synthetic participants when you are still narrowing down: many concepts, many questions, no budget or time to recruit for each. Use AI-moderated interviews with real people when you need evidence you can stand behind for a decision. The two work well in sequence. Synthetic interviews tell you where to look, and real interviews confirm what you found.
Start with a free project, or walk through Candor with us first.