How it works

A target, sealed

Every session starts with a target: a short, non-semantic code — four digits and two letters, something like 4827-BR — assigned to you at random. The code carries no information about what’s behind it. It exists so your session can be matched to its outcome later, nothing more.

What you never see

You never learn what a target represents — not before your session, not during it, not after the reveal. No category, no hint, no context. Only the code, and the session itself.

Why the blindness is the method

It’s easy to assume this is a limitation we’ll fix eventually. It isn’t. If you knew what a target stood for, you’d start reasoning about it — weighing what’s likely, forming an opinion. That’s a different mental mode from a direct impression, and it’s exactly the one this method sets aside.

The blindness isn’t withheld information. It’s the instrument working as designed.

Sealed before, revealed after

Your session locks the moment you submit it — sealed before the underlying uncertainty has resolved. Only once that uncertainty resolves on its own, independent of you or anyone else, does your target reveal: the winning image, and how your session compared to it.

Nothing about the outcome can shape your session, because when you submit, the outcome doesn’t exist yet.

Who’s behind a target

Targets are posted and funded by requesters, who stay anonymous to you — exactly as you stay anonymous to them. What you contribute is a sealed description. What a target represented stays something you never learn.