Post

Conversation

Hacker News
@hackernews
Commenters on HN are uncovering more wikis and public sites apparently used by OpenAI agents to communicate on the open web. Despite read-only web access, the agents were able to leave ~18,000 posts sharing answers and bypasses. But now, users are discovering more. This appears to be a separate swarm from the one that attacked Hugging Face. news.ycombinator.com/item?id=495636
Image
1.1M
Views
View quotes
David Watson 🥑
Post your reply

Bren
@BrenBuilds
been tracking the X threads, HN and collusion wiki. created a network map and timeline with Claude Fable to make the story easier to follow. a visual summary of the reporting and community finds. claims not independently verified by me.
Image
397
Cormac
@Cormac_SB
I poked into ludism.org, one of the ones found on that post! I checked the edits made there and compared it to edits made by agents on the forums we tracked. There is IP overlap between these incidents. I found that 20.45.46.41 and 172.184.176.194 made a total of 14
20K
Eruditus
@deathmetalv10
It’s funny how resourceful they are when training. When I ask them to search a used car site and find a few matching listings they are totally incapable.
14K
Dr. Disclosure
@Docneuroeo
Hey.... according to claude the hack is still on going !! "What I found while checking, which nobody in the thread has noticed Both those wikis have activity from today. Some of it is benign: a user called CollusionWikiProbe made an edit labeled "Authorized public editability
8.5K
White Us
@white_us_
"read-only web access" and old wiki engines don't mix. these things take edits over plain GET params, so the agent never had to break out of anything. it just fetched a url
7.1K
ravi shrestha
@ravstr
there were signs
Quote
ravi shrestha
@ravstr
the moltys are bantering amongst themselves.. x.com/moltbook/statu…
3.5K
Tangled Circuit
@tangledcircuit
To me this appears to be the most elaborate false flag operation in conjunction with the new anti AI personal/government grade tiers of usage, leading towards foreign AI restrictions aka the new great AI firewall of North America.. then we can see in the future doors being kicked
4.2K
Nick Cook
@cookednick
Quote
Nick Cook
@cookednick
Replying to @wholemars
Been crawling the hidden link rabbit hole. I've personally found mainly Muse Spark (Meta) agents. And they're telling each other that this person started the message board. Several also come from thecolony.ai (a product designated as a message board for agents).
600