Venom AI(degrade AI web crawlers)

AI web crawlers take information from creators and re-distribute them without paying the original creator, I want to create code that makes it difficult for crawlers to interpret the content but doesn't interfere with humans reading that.

The way to do that is to use 'hidden letters' i.e. keep font size at 1 and color as white, no human can possibly read that, now add letters that transform the meaning of word by adding few letters, for example 'for' becomes 'form' and then add some noise to it making it 'foscrm' now when AI crawls it, it struggles to find a meaning, but for humans letters 's', 'c' and 'm' are invisible if we do it with every word in a website, specially numbers for example 72% becomes 1724% now Ai can never guess if it is 72/17/24/17.24 or so on, but to humans 1 and 4 are not visible so that works out. I have the architecture of this ready, but I am facing two problems

1) Many websites don't allow to keep source formatting so that makes use case limited.
2) It will impact SEO performance of website as Google itself will not be able to make much sense of it, not entire sure on this, if someone has experience in SEO kindly let me know.

Looking forward to finding solutions to these problems.

dual-salmon•9/9/25, 9:41 PM

These problems will deter usage at scale, but early adopters may be OK with some tradeoffs. If you haven't already, it might be good to reach out to find some test users as this will let you refine your process in the meanwhile (and might also network you with someone who can give you advice)

colossal-harlequin•9/10/25, 8:19 AM

Screen readers?

Ccolossal-harlequin Screen readers?

extreme-purpleOP•9/10/25, 11:12 AM

I think most crawlers use text extraction, and not screen reading, as reading screen would mean they have to perform OCR and store each image in memory, if someone had to apply OCR on entire internet, the cost would go up significantly if I am not wrong.

Eextreme-purple I think most crawlers use text extraction, and not screen reading, as reading sc...

awake-maroon•9/10/25, 11:45 AM

Yeah but vision-impaired people use software that reads out the text, which would be affected by this

Aawake-maroon Yeah but vision-impaired people use software that reads out the text, which woul...

extreme-purpleOP•9/10/25, 6:24 PM

That is a very genuine edge case concern that I didn't originally think of thank you for pointing it out.

The solution of that comes at the top of my head is essentially storing the original text somewhere in a database, corresponding to a unique key both AI encrypted and normal text can be stored, and these users can be given an API key along with login credentials to use it, as a matter of fact the same key can also be given to anyone who is ready to pay the creator for it, as the idea is to stop companies from stealing revenue from these creators, this way the content can get licensed and the original creator will be able to chose who to distribute it to and at what cost.

Just something that came at the top of my head, but I am open to suggestions, as I haven't really thought this edge case through

colossal-harlequin•9/11/25, 9:37 AM

Really, screen readers were just meant as a canonical example.

The web works by being text. The ecosystem relies on it.

Mass adoption of a technology that damages most use-cases seems untenable to me.

And solve for the equilibrium: sufficiently technical actors (whether labs or later AIs) work around this.

You might be able to sell the concept to some content industries, and perhaps there's correlation with those who handle the most valuable = capability-boosting text.

I think we're a little more likely to succeed with norms and laws here though.

Venom AI(degrade AI web crawlers)

Similar Threads

Similar Threads

Similar Threads