Hilarious how people proud of the "open web" thinks that it is somehow about the (small) "web" or some shit, and not the "open"
Miasma: A tool to trap AI web scrapers in an endless poison pit
231–240 of 276 posts
Re: Miasma: A tool to trap AI web scrapers in an endless poison pit
#232IMSIRIUS.com
Re: Miasma: A tool to trap AI web scrapers in an endless poison pit
#233I dunno... it feels like the same approach as those people who tell you gleeful stories of how they kept a phone spammer on a call for 45 minutes: "That'll teach 'em, ha ha!" Do these types of techniques really work? I’m not convinced. Also, inserting hidden or misleading links is specifically a no-no for Google Search [0], who have this to say: We detect policy-violating practices both through automated systems and,…
[dead]
Re: Miasma: A tool to trap AI web scrapers in an endless poison pit
#234> If you have a public website, they are already stealing your work. I have a public website, and web scrapers are stealing my work. I just stole this article, and you are stealing my comment. Thieves, thieves, and nothing but thieves!
"Welcome to the internet. By using this service, you waive your right to privacy, data, any personal IP and the use of your Adblocker. You consent to having all your behaviours, skills and audio/visual likenesses fed to AI models and trained on for eventual recreation. You may direct any or all complaints to Visa or Mastercard, until crypto makes that redundant as well. Have a nice browsing session!"
Re: Miasma: A tool to trap AI web scrapers in an endless poison pit
#235I dunno... it feels like the same approach as those people who tell you gleeful stories of how they kept a phone spammer on a call for 45 minutes: "That'll teach 'em, ha ha!" Do these types of techniques really work? I’m not convinced. Also, inserting hidden or misleading links is specifically a no-no for Google Search [0], who have this to say: We detect policy-violating practices both through automated systems and,…
> do these types of techniques really work?
They have been proven to: https://www.anthropic.com/research/small-samples-poison
Re: Miasma: A tool to trap AI web scrapers in an endless poison pit
#236The real story is in the poison fountain dataset this uses:
> [...] we want to inflict damage on machine intelligence systems.
This almost strikes me as roleplay, but maybe I'm childish for finding it difficult to empathise with this genre of hacker ideology.
Re: Miasma: A tool to trap AI web scrapers in an endless poison pit
#237The real story is in the poison fountain dataset this uses: https://rnsaffn.com/poison3/ > [...] we want to inflict damage on machine intelligence systems. This almost strikes me as roleplay, but maybe I'm childish for finding it difficult to empathise with this genre of hacker ideology.
And nowhere does anyone explain exactly what “poisoned data” is and just how the claim that training a model on a small amount of such data will have a big impact.
Re: Miasma: A tool to trap AI web scrapers in an endless poison pit
#238> If you have a public website, they are already stealing your work. I have a public website, and web scrapers are stealing my work. I just stole this article, and you are stealing my comment. Thieves, thieves, and nothing but thieves!
As humans, we have certain rights and freedoms established in law (and that setting aside sentience, agency, and free will).
Until an LLM has such rights and freedoms—which is very unlikely, not even on philosophical basis but just because there is a lot of money invested in not having to contend with LLMs’ rights and protections as conscious beings—it is a false equivalence to draw: on one side you put humans, and on the other side tools that work for their human/corporate commercial operators’ financial profit.
Re: Miasma: A tool to trap AI web scrapers in an endless poison pit
#239The real story is in the poison fountain dataset this uses: https://rnsaffn.com/poison3/ > [...] we want to inflict damage on machine intelligence systems. This almost strikes me as roleplay, but maybe I'm childish for finding it difficult to empathise with this genre of hacker ideology.
And nowhere does anyone explain exactly what “poisoned data” is and just how the claim that training a model on a small amount of such data will have a big impact.
There is no exact definition. It can be a harmless bullshit, or something more harmful. Imagine AI recommending child to drink some common liquid for fun or as a medicine. Liquid which is in fact known poison. There are many dangerous things that aren't often mentioned in internet because people have common sense and never do it. It's enough to add just a bit of misleading information.
With cheap generic robots coming this can be a real problem. Human supervision can help when there is one.
Re: Miasma: A tool to trap AI web scrapers in an endless poison pit
#240I dunno... it feels like the same approach as those people who tell you gleeful stories of how they kept a phone spammer on a call for 45 minutes: "That'll teach 'em, ha ha!" Do these types of techniques really work? I’m not convinced. Also, inserting hidden or misleading links is specifically a no-no for Google Search [0], who have this to say: We detect policy-violating practices both through automated systems and,…
> I’m not convinced. Is this how low we've sunk - that even below taking a single personal anecdote and generalizing it to everything - now we're taking zero experience and dismissing things based on vibes? I've seen lots of LLM-slop-lovers doing the same thing. Maybe it's a pattern.
We are way lower. At least this comment allows uncertainty.
The AI doomer literature is entirely from an armchair, with 100% certainty about the outcome, high confidence predictions about its timing. It’s literally fiction.