The private AI journal that builds on what you actually write
Say it before you edit it. Then let it accumulate until something you keep circling comes into view.
What the evidence states mean
Most apps flatten everything you write into one pile, so a passing worry sits next to a decision you have already made. ThoughtStream Lab keeps track of how much weight a thought has actually earned. Every entry carries an evidence state, and it only moves when there is something behind the move.
The assistant proposes the state. You decide. Nothing is treated as true because an AI said so, and nothing you wrote is ever rewritten, tightened up or corrected in place.
- CapturedA raw thought just recorded. No judgement yet.
- ObservedSomething you noticed happening, stated as a fact.
- RecurringThis keeps coming back across multiple thoughts.
- InferredA hypothesis or interpretation — not yet confirmed.
- ConfirmedBacked by enough evidence to treat as true.
- RejectedRuled out by contradicting evidence.
- ArchivedSet aside — no longer active, but kept for the record.
What it reads across your entries for
- What keeps coming back. A theme you return to without naming it, pulled out of your own entries rather than a category you picked.
- Where you contradict yourself. Two entries that cannot both be true, set side by side so you can decide which one still stands.
- What connects. Ideas from different weeks that turned out to be about the same thing.
When something has earned enough weight, you turn it into a commitment and review it each week to see what changed. The thought stays in your ledger exactly as you wrote it, and everything the assistant makes of it stays beside it, in its own place.
Private by design
Your ledger belongs to your account alone. What you write is encrypted at rest, and it is never read to train anything or handed to anyone else. Read the privacy policy.