this post was submitted on 28 Aug 2023
477 points (96.1% liked)
General Discussion
13297 readers
3 users here now
Welcome to Lemmy.World General!
This is a community for general discussion where you can get your bearings in the fediverse. Discuss topics & ask questions that don't seem to fit in any other community, or don't have an active community yet.
πͺ About Lemmy World
π§ Finding Communities
Feel free to ask here or over in: !lemmy411@lemmy.ca!
Also keep an eye on:
- !newcommunities@lemmy.world
- !communitypromo@lemmy.ca
- !new_communities@mander.xyz
- !communityspotlight@lemmy.world
- !wowthislemmyexists@lemmy.ca!
For more involved tools to find communities to join: check out Lemmyverse!
π¬ Additional Discussion Focused Communities:
- !actual_discussion@lemmy.ca - Note this is for more serious discussions.
- !casualconversation@lemm.ee - The opposite of the above, for more laidback chat!
- !letstalkaboutgames@feddit.uk - Into video games? Here's a place to discuss them!
- !movies@lemm.ee - Watched a movie and wanna talk to others about it? Here's a place to do so!
- !politicaldiscussion@lemmy.world - Want to talk politics apart from political news? Here's a community for that!
Rules and Policies
Remember, Lemmy World rules also apply here.
0. See: Rules for Users.
- No bigotry: including racism, sexism, homophobia, transphobia, or xenophobia.
- Be respectful. Everyone should feel welcome here.
- Be thoughtful and helpful: even with βsillyβ questions. The world wonβt be made better by dismissive comments to others on Lemmy.
- Link posts should include some context/opinion in the body text when the title is unaltered, or be titled to encourage discussion.
- Posts concerning other instances' activity/decisions are better suited to !fediverse@lemmy.world or !lemmydrama@lemmy.world communities.
- No Ads/Spamming.
- No NSFW content.
founded 2 years ago
MODERATORS
you are viewing a single comment's thread
view the rest of the comments
view the rest of the comments
I wonder if a bot using AI image recognition would be feasible. Train it on CP and similar awful stuff and have it auto-flag posts that fit the bill for moderator removal. The problem would be sourcing the training material and finding people willing to expose themselves to what it flags.
Good luck with explaining storing of that material to law enforcement agencies
IIRC Stable Horde already has an open source model for detecting it
The best thing to do would be to train it at first by having it trained on live posts as human moderators flag CSAM, then once it's trained up, it can start auto-flagging posts, with human mods checking. Don't keep the CSAM material, just train the neural net, then delete.
This should be doable without storing CSAM for any longer than it takes to catch it and remove it.
FYI there's a considerable problem with high precision ML recognition being "reversible" into generative algorithms
Honestly just auto-remove it and let users appeal
How will you detect automatically which photo to remove and which not to?
Something this already exists and is used by google and law enforcement agencies.