The AI that interviews you: why questions beat dictation

You tried dictating a post once. You talked for four minutes, looked at the transcript, and closed the app. It read like someone talking — because it was. Loops, half-sentences, three versions of the same point.

The problem isn’t that you can’t talk your way to a draft. It’s that a monologue is the wrong format. Nobody rambles their way to a sharp take. But almost everyone gets sharp when a good interviewer pushes back.

Journalists have known this forever. The subject doesn’t write the profile. They answer questions, and the questions do the structuring.

Run the interview method manually

You can do this today with no special tools:

1. Record a two-minute ramble on the idea, in Voice Memos or anything else.
2. Transcribe it. Apple’s built-in transcription or any dictation tool works.
3. Read it and write down the three questions a skeptical friend would ask. This is the hard part: you have to interrogate yourself.
4. Record yourself answering those questions, one at a time.
5. Transcribe again, then assemble. Your answers are the body of the draft. The strongest answer is usually your opening line.

The method works. The catch is step 3: writing good questions about your own thinking is exactly what your brain avoids doing. You already agree with yourself. And the loop is 30–45 minutes of app-juggling, so most people do it once.

The collapsed version

Xtraktr runs that loop as one conversation. You say the rough version out loud, and it asks one question at a time about what you just said — where your claim is vague, what your example actually shows, where a skeptic would push. You answer, it asks the next one.

You can steer it, too. Tell it to Deepen when the interesting part is under the claim, Clarify when you catch yourself being fuzzy, Synthesize when you’ve talked enough and want the thread. When the idea is finally solid, you turn the conversation into an outline or a summary assembled from your words, not generated ones. From there, a blog post, a LinkedIn post, or a newsletter issue is mostly done.

One more difference from every dictation tool: the interview happens on the phone. The transcription, the questions, and the drafts all run on the iPhone’s own chip. Your half-formed thoughts never touch our servers, which matters when the idea you’re working out isn’t ready for anyone else yet. Here’s what on-device actually means.

Why this beats “AI, write me a post”

You could paste a topic into a chatbot and get a post in ten seconds. It will be smooth, plausible, and interchangeable with everyone else’s. It’s built from the average of the internet, not from you.

An interview can’t produce that, structurally. Every sentence in the output traces back to something you said. Your examples, your phrasing, your actual position. The AI’s job is only to ask and to structure, so the thinking stays yours. That’s the difference between sounding like an expert and having your expertise extracted.