3 ms·
I love this! But won't the machine easily pick up on this?
by pedrogpimenta 4mo ago
I love this! But won't the machine easily pick up on this?
- cwnyth 4mo agoAs they said in the comment you replied to: "Note that a sufficiently prompted AI agent can definitely read this, so it's not meant to be cryptographically sound, more just unfriendly to the common AI reader!"
- billtarbell 4mo agoIt's actually not really easy for AI, without the agent doing some actual coding itself to reverse engineer the font file, or to take screenshots at different variable font intervals to zone in on the "focused" version of the variable font. All of that being said, the intention (beyond just having fun creating it) was to make it AI "unfriendly" so AI bots doing broad quick reads of it are going to be left with gobbledegook encoded characters.
- anon291 4mo agoMost llms can equally engage with text in picture form as text in token form. In fact my initial research on this (later corroborated by actual published papers) indicate that this is a cheap way to save on tokens.
- billtarbell 4mo agoOh interesting and good to know on the token savings with this technique. My test with claude had it use vision and then programmatically test different variable font input variables (mimicking the user scrub interaction) until it was able to OCR it.
- anon291 4mo agoI mean I can't know for sure but I'm pretty sure that by the time the upper layers of the network are reached the lower level networks have already transformed the image tiles into proper position encoded embeddings of the tokens in the words in the image. That would be my operating assumption at least.
- billtarbell 4mo agoThe encoded data and the font file go over the network. The font file is executed locally on the readers machine.
- anon291 4mo agoMy point is you can just render the file and send to AI...