I built Don’t Ask Me — an AI chat app that does the opposite of what most AIs do. Instead of giving you instant answers, it helps you reason things out, reflect, and build your own conclusions.
The idea came from noticing how easy it’s become to outsource thinking to LLMs. I wanted to experiment with a tool that doesn’t take over your thinking — but guides it.
How it works:
You ask a question or describe a problem.
Instead of replying with a final answer, the AI asks you questions back.
It helps you unpack assumptions, find contradictions, and explore perspectives.
The goal isn’t to “get an answer” — it’s to think better.
Why: I wanted something between journaling, Socratic dialogue, and ChatGPT — something that builds resilience of thought instead of dependency.
Note: This is very much an MVP — just the first working version of the concept. There’s still rough edges, limited depth, and no fancy onboarding yet. I’m more interested in validating whether this type of AI interaction feels valuable at all before polishing it.
I’d genuinely like your thoughts on:
Whether the “refuse to answer” approach feels helpful or just frustrating
The UX of being challenged instead of spoon-fed
Any ideas for making the reasoning scaffolding smarter or less repetitive