- cross-posted to:
- [email protected]
- cross-posted to:
- [email protected]
Can we poison our posts by putting a nonsense “signature” at the end of each of them?
I don’t see why everyone’s surprised about this. The Fediverse is running on ActivityPub, an open protocol whose purpose is to broadcast the content we post here to anyone who wants it. Of course it’s being used to train AI, why wouldn’t it?
Except iirc, they aren’t scraping “properly” (read: efficiently at least, setting aside morality for the sake of discussing this component in isolation), and are causing traffic troubles. If only they took the time to install an actual instance themselves then nobody would care in the slightest (again, ignoring the morality part, for now).
TLDR: they are being dicks about it, bc offering everything we have for free is not enough for them.
of all the scrapers we see, the requests identified as originating from Meta seem to be well behaved overall. they appear to (mostly) be respecting robots.txt where present and their request volume to Lemmy.World is only averaging slightly above 5 requests per minute over the last 2 weeks. they also don’t spoof their user agents to pretend to be web browsers, or at least I have not seen credible accusations of this happening.
Thank you for sharing that data 😊
i mean, that’s exactly what they did with threads, and many instances defederated from it because they didn’t want to have their data scraped by meta
But if they do it the “proper” way, they won’t be able to grab the data if instances defederate from them, right? And that’s what the majority of instances will do.
Assuming you know which instances are the ones they’re collecting data from. It could be any instance.
You are absolutely correct there, in that hypothetical scenario if they were to attempt to hide their traffic among normal instance activities.
To add a bit more detail to my previous answer, there were some prior discussions about this topic, citing some of the most popular instances of the entire Threadiverse having been targeted by their normal DDOS-like approach:
Good.
I do not know enough about the ActivityPub protocol to answer that. I did think that federation at least used to be the default many years ago but aren’t sure about the current status of that. Indeed detection and subsequent blocking will always be the cat and mouse game that is played but use of ActivityPub might at least delay the former part? And how would anyone find out, compared to e.g. if not a single person household then at least a small community instance just wanting to pull down all the content across the Fediverse to read up on?
At this point, I appreciate that anyone can scrape it. Not just Reddit or Meta exclusively, but any start up that’s wants to compete. Sure, meta and the biggies have an easier time of it, but at least they don’t get it all only for themselves.
That doesnt necessarily mean that training AI on this data is legal. Especially when multiple of these instances had legal documents in place specifically forbidding this kind of use.
There are some lawsuits in motion about this and the early signs are that it is indeed legal. For example, in Kadrey et al v. Meta the judge issued a summary judgment that training an AI on books was “highly transformative” and fell under fair use, and similarly in Bartz, Graeber and Johnson v. Anthropic the judge ruled that training an AI on books was fair use. I always expected this would be the case since an AI model does not literally contain the training material it was trained on, it learns patterns from the training material but that’s not the same as the literal expression of the training material. Since the training material isn’t being copied there’s nothing for copyright to restrict here.
it isn’t about surprise silly goose its about moving the interaction from a suspected unknown to a known interaction in our collective threat models
silly goose
Copy and pasting my own list from here
List of instance
beehaw.org furry.engineer ibe.social fediworld.de framatube.org trailers.ddigest.com nrw.social lemmynsfw.com video.hardlimit.com digitalcourage.social xn--baw-joa.social tube.kockatoo.org equestria.social wisskomm.social social.anoxinon.de freiburg.social toobnix.org toot.bike mstdn.lalafell.org peertube.linuxrocks.online social.rebellion.global mastodon.cipherbliss.com social.sdf.org corteximplant.com typo.social www.404media.co mastodon.ml video.liberta.vip tilvids.com todon.eu hessen.social digipres.club shigusegubu.club mastodon.me.uk zdf.social mastodon.sdf.org spore.social kolektiva.media gruene.social share.tube nso.group mastouille.fr masto.es vivaldi.com literatur.social mstdn.mx kirche.social mastodon.hams.social federation.network lile.cl todon.nl betweenthelions.link ipv6.social linuxrocks.online peertube.otakufarms.com pawb.social mastodon-belgium.be jasette.facil.services machteburch.social mastodont.cat mastodon.eus eupolicy.social social.bau-ha.us toot.berlin amicale.net hexbear.net mastodon.bida.im reddthat.com shelter.moe mastodon.nl dju.social bonn.social mstdn.chrisalemany.ca social.sciences.re tldr.nettime.org lemy.lol climatejustice.social rollenspiel.social mastodon.org.uk social.kyiv.dcomm.net.ua pouet.chapril.org ecoevo.social social.politicaconciencia.org darmstadt.social peertube.tv lemmus.org libretooth.gr hackers.town tooter.social anarchism.space diode.zone video.infosec.exchange mastodon.thirring.org aussie.zone social.bund.de apobangpo.space shitpost.cloud berlin.social toot.aquilenet.fr social.beachcom.org lemmygrad.ml mastodon.radio nerdculture.de programming.dev decayable.ink kafeneio.social functional.cafe things.uk fuzzies.wtf diaspodon.fr dalek.zone sunbeam.city tooting.ch fediscience.org mastodon.tetaneutral.net social.librem.one im-in.space lemmy.sdf.org legal.social post.lurk.org mastodon.uy noc.social tube.pol.social lemmy.ml don.linxx.net infosec.pub kolektiva.social masto.bike furries.club zhub.link lemmy.world openbiblio.social mastodon.zaclys.com mamot.fr clacks.link discuss.tchncs.de cyberplace.social graz.social pl.kitsunemimi.club mastodonczech.cz masto.nobigtech.es hostux.social pawb.fun mastodon.trueten.de norden.social systemli.social mander.xyz ciberlandia.pt woem.men sopuli.xyz lemmy.ca feddit.uk
off-topic but wow, it’s great to see so many lemmy instances up and running 🥰
it really looks like we’re well on the way to hitting critical mass
deleted by creator
# Convert PDF to text pdftotext Meta_Leaked_List.pdf meta.txt # Fetch instances feddit.uk federates with curl https://feddit.uk/api/v3/federated_instances | jq -r .federated_instances.linked.[].domain > instances # Get the lines that overlap between the two files grep -xFf instances meta.txt
Sorry lol, turns out my client just didn’t render the post properly with the additional link to the file!
I see that shitposter.club is on the list. Good to know they’re using only the highest-quality training material.
I tried to visit but their security certificate is expired. Are they still a legit site?
Moved to shitposter.world according to their site with the expired cert, but I haven’t seen as much on fedi from the new domain as I used to from the old one.
This is only a loosely related thought, but are there any new foss licenses or anything that prohibit ai usage? I know it’ll be ignored but it feels like explicitly disallowing things could be important in opening the door to successful legal challenges to ai scraping and theft…
Case law is still pretty young in this area, but it’s looking like there’s nothing actually against copyright about the training of AI on copyrighted content. It’s not something that a license can restrict because the trainers can simply reject the license and carry on training under the basics of what the law allows them to do anyway.
Open source licenses only have power because they grant permissions that people normally wouldn’t have and put conditions on those permissions. If you don’t need those permissions then you don’t have to be bound by those conditions.
Ahhh, that sucks ass :(
Thank you for expanding my understanding of the problem!
Those tasteless frauds!