LLMs can unmask pseudonymous users at scale with surprising accuracy

https://arstechnica.com/security/2026/03/llms-can-unmask-pseudonymous-users-at-scale-with-surprising-accuracy/

Burner accounts on social media sites can increasingly be analyzed to identify the pseudonymous users who post to them using AI in research that has far-reaching consequences for privacy on the Internet, researchers said.

34 points · 7 comments · view on lemmy.world

7 Comments

refalo@programming.dev · 18 pts · 166d

Only if your OPSEC was already bad... LLMs aren't magical, they're just piecing together information you already gave away freely.

Also the authors of this paper refuse to show their work "for safety."

village604@adultswim.fan · 8 pts · 166d (2 replies)

I've actually considered using a local AI to rewrite my comments for me to defend against this.

Bibip@programming.dev · 4 pts · 166d

Same. I think it's ghastly that an internet where individuals are behaving safely is an internet where every single comment is the same sterile clank.

GeeDubHayduke@lemmy.dbzer0.com · 2 pts · 165d

Using AI to hide from AI? Makes sense.

altphoto@lemmy.today · 7 pts · 166d

Create an account every week. If the place doesn't allow that then it's clearly time to move on. I'm finding it less rewarding to continue a single account than to start new ones. Its fun. Plus whoever is left will not need AI to be identified easily.

Skyrmir@lemmy.world · 1 pts · 166d

Use an AI bot that creates an account, posts in a nondescript fashion, and tracks responses. Then the whole internet becomes anonymous soup.

greatwhitebuffalo41@slrpnk.net · 1 pts · 166d

Back to lurking you say? Ugh I guess... But realistically, this account is tied to my Reddit account. If I'm gonna do weird stuff, it's not gonna be here.