upvote
> a screenshot

The text is a video. Every frame contain random dots, so an individual frame by itself doesn't contain the intended message

This "font" exploits the fact that current-gen frontier models will process video one frame at time, but each frame is noise, so by looking at frames in isolation doesn't reveal anything

Then, they add a hidden message to each frame just so that the agent report something and stop trying (because if the agent tried to correlate between the frames, they could discover the trick)

But if you pass just a frame, there is no message. Just the noise plus the decoy

reply
If you take a frame you see it's neither random nor dots:

https://i.imgur.com/CgtyGjl.png

From a single frame you can definitely identify boundaries because the dots are sliding and get truncated.

reply
They are the boundaries of the decoy, I think. I can sort of make the decoy out in a screenshot.
reply
Exactly. It's a good idea, badly executed.
reply
Maybe not so well explained, by picking the default intended text presented to be the same as the decoy text. It took me also some time to realise what was going on, but the execution is fine otherwise.

So there are two texts, one decoy (which you can barely see in a single frame but becomes more clear if you average between frames) and an actual text, which disappears in single frames or averaged ones.

reply
Just to add some drama - this feels like a perfect competition between humans & bots!
reply
"Content not available in your country" - obviously working well.
reply
What did you expect from a screenshot of obvious noise? The only thing that makes the text readable is the motion.

EDIT: On second look, the static screenshot does say "WRITTEN IN GHOST FONT".

reply
deleted
reply
> I posted a screenshot of static white noise to AI

HackerNews never disappoints

reply