811
[AskLemmy Meta]: Should bot posts be banned in AskLemmy.world? (lemmy.world)
submitted 4 days ago by [M] to c/asklemmy@lemmy.world
126 comments fedilink hide all child comments
 

The recent post by an AI agent didn't technically break any of the rules, but it feels like it's against the intent of the community. Upvote for YES. Downvote for NO.

you are viewing a single comment's thread
view the rest of the comments
[–] 81 points 4 days ago* (last edited 4 days ago) (5 children)

Of course they should be banned. Question is how?

Captchas don't stop all bots anymore. Would probably prevent most, but then it's a question of how much to annoy legitimate users.

It's not like I'm willing to solve one each time I post (human here, BTW). But if it's a one-time thing at registration, not much stops a human from signing up then letting their bot go wild.

And that's before even thinking about the federation problem. If most instances are welcome to interact, how to prevent small bot instances? You could allow-list, but that harms legit small or single user instances.

Edit: just read that post where the agent was upfront about being an AI trying to scam money. And I'd say that shouldn't be allowed. The ones that aren't honest about it bother me more tho, and I suspect there are a lot more of those posts than we realize.

  • permalink
  • fedilink
  • source
  • hideshow 10 child comments
    •
  • [–] 26 points 4 days ago (2 children)

    Most anti-ai tech isn't captcha anymore. Most captcha tech was explicitly for training image recognition for AI, and for any that are effective against AI there are many services that will allow you to buy "solves" (send the captcha to someone in a third world country to solve for you) for like a dollar for 100.

    Most methods now are either proof of work (make the machine connecting do some computations that take like half a sec for a real user but is costly enough to deter mass AI scraping), traps like having hidden links that no real user would click but that an AI crawling the file directly instead of browsing visually will click and send them into an infinitely generated maze of passable nonsense pages, or complicated behavioral and browser fingerprint analysis (all the click here to verify you're human boxes that don't make you do anything else).

    A number of Lemmy instances, like the one I'm on dbzer0, implement one or more of these features already.

    A secondary issue is that these methods are mostly meant to stop AI web scrapers, not AI bot posters. Since most Lemmy instances allow simpler non-LLM bots to make posts, like reddit reposting, or posting contents from RSS feeds, there's no automated way to stop LLMs except trying to stop all bots (assuming they all are using the API, and the LLMs don't communicate any self identifying metadata).

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 4 child comments
    •
  • [–] 2 points 3 days ago*

    there’s no automated way to stop LLMs except trying to stop all bots (assuming they all are using the API, and the LLMs don’t communicate any self identifying metadata).

    there's no good way to do this due to federation: that means there's always ways that bots just make their own instance and post from there.

  • permalink
  • fedilink
  • source
  • parent
  • [–] 10 points 4 days ago (1 child)

    In this case they actually correctly flagged themselves as a bot, so just automatically remove posts from users marked as a bot. World is already behind cloudflare so that obviously didn't stop it.

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 2 child comments
    •
  • [–] 7 points 4 days ago

    Of course they should be banned. Question is how?

    Lemmy.world already uses CL services. Cloudflare has a vested interest in being the most accurate game in town for AI bot recognition (they intend to swap to a usage pricing model) and has been laying ground since early this year.

    As of July 1, they made new bot options available to their clients, with more to come (https://blog.cloudflare.com/content-independence-day-ai-options/).

    I presume, .world administration would turn to those and/or use them in conjunction with other discouragement tools.

  • permalink
  • fedilink
  • source
  • parent
  • [–] 2 points 3 days ago

    They are already using Fediseer to filter out instances that are vulnerable or made for spam. They ahould implement Cap on the website which detects headless/automated browsers and uses PoW to slow down very stealthy bots, then fall back to PoW combined with a 4chan-style captcha for mobile apps.

    They can then use Shieldstral to detect spam, harrassment, and anything illegal which gets through from very stealthy bots and humans.

  • permalink
  • fedilink
  • source
  • parent