Start here
The free tool below runs entirely in this browser — no account, nothing stored or transmitted. Redact anything you would not want kept before you type it.
Open the companion tool →Why intent matters more than hours
An app used an hour a night as a writing partner and an app used an hour a night as the only voice that answers are the same number and completely different situations. Time is the measure everyone reaches for and the one that tells you least.
Naming the need is what makes the adjustment possible. If the app is standing in for company, the answer might be one phone call a week. If it is standing in for a thinking partner, the answer might be nothing at all.
The protocol
- Write down the specific thing you wanted the first time you opened it. Be concrete: company, calm, a deadline, an argument you were rehearsing.
- Write down what you actually get now, in the same concrete way. Include what it does well — a competent answer deserves to be recorded.
- Note the gap, if there is one. A gap is information, not a failure.
- Choose one adjustment you control — a time of day, a substitute activity, one human conversation — rather than a rule about the app.
- Date the note and look at it in a few weeks. Adjust again if needed; the point is the direction, not obedience.
What the tool does
Compare what you wanted from the app with what you actually get, and write down one adjustment you control. It scores nothing and stores nothing beyond this page.
Limit first: This is a reflection aid, not a diagnosis, and not a judgment about your relationship with any app. You are the only person who can say what you were looking for.
Open the companion tool →One human next step
If what you are getting is company and what you want is people, send one message to one person this week. That is the whole next step, and it does not require giving up the app.
Research log and safety checks
AI-assist path: not applicable — the comparison is between the reader's own two statements; a model would only restate them, and its voice would blur whose judgment is being recorded. §6.10: not a frontier-pace topic.
§0.5 protection result: category is chatbot-dependency-check; crisis-adjacent: no; fear/urgency/scarcity toolkit used: no. The protective function is free and the paid feature is convenience only.