this post was submitted on 09 Mar 2026
135 points (93.5% liked)

Technology

82460 readers
3968 users here now

This is a most excellent place for technology news and articles.

Our Rules

  1. Follow the lemmy.world rules.
  2. Only tech related news or articles.
  3. Be excellent to each other!
  4. Mod approved content bots can post up to 10 articles per day.
  5. Threads asking for personal tech support may be deleted.
  6. Politics threads may be removed.
  7. No memes allowed as posts, OK to post as comments.
  8. Only approved bots from the list below, this includes using AI responses and summaries. To ask if your bot can be added please contact a mod.
  9. Check for duplicates before posting, duplicates may be removed
  10. Accounts 7 days and younger will have their posts automatically removed.

Approved Bots

founded 2 years ago
MODERATORS
L3s@lemmy.world
enu@lemmy.world
technopagan@lemmy.world
L4s@lemmy.world
L3s@hackingne.ws
135
AI allows hackers to identify anonymous social media accounts, study finds (www.theguardian.com)
submitted 1 day ago by HellsBelle@sh.itjust.works to c/technology@lemmy.world
9 comments fedilink hide all child comments
 

AI has made it vastly easier for malicious hackers to identify anonymous social media accounts, a new study has warned.

In most test scenarios, large language models (LLMs) – the technology behind platforms such as ChatGPT – successfully matched anonymous online users with their actual identities on other platforms, based on the information they posted.

The AI researchers Simon Lermen and Daniel Paleka said LLMs make it cost effective to perform sophisticated privacy attacks, forcing a “fundamental reassessment of what can be considered private online”.

In their experiment, the researchers fed anonymous accounts into an AI, and got it to scrape all the information it could. They gave a hypothetical example of a user talking about struggling at school, and walking their dog Biscuit through a “Dolores park”.

In that hypothetical case, the AI then searched elsewhere for those details and matched @anon_user42 to the known identity with a high degree of confidence.

Study link - https://arxiv.org/abs/2602.16800

all 10 comments
sorted by: hot top controversial new old
[–] Scubus@sh.itjust.works 4 points 8 hours ago

Identify Deez nuts

[–] silverneedle@lemmy.ca 9 points 15 hours ago (1 children)

I've said it before and I will reiterate it a thousand times if need be. AI based analysis, particularly LLM based analysis, will strictly lead to bogus results. The utility for this is intimidation and culling (!) of the have-nots.

Imagine entering a country and being flagged for antidemocratic rhetoric because the computer said so. It doesn't matter if you said it or not, the machine has a claimed .01% error rate. Furthermore, it doesn't matter if this error rate is correct or not, because how the machine got its results is a process that is impossible to pry open. Anyone who manages an algorithm can plant something and it's very difficult to know who because the type of person to be able to do so will bring with them the precaution to wipe their tracks.

That said, you can mark my words. You'll be seeing a lot more of this type of brutalisation in the future. The target is anyone who is among the patsies of state spun narratives, trans people, immigrants, specific nationalities, yadda-yadda.

[–] surewhynotlem@lemmy.world 37 points 1 day ago (1 children)

My girlfriend used to spend weeks stalking through years old comments to unearth info about a person. Now she's being replaced by AI.

[–] gapa@feddit.nu 1 points 16 hours ago

This is why we have to stop AI.

[–] Kolanaki@pawb.social 3 points 15 hours ago* (last edited 15 hours ago)

This is only possible if you have other SM where you're not anonymous; which could have just as easily been linked to your anonymous accounts without the use of AI. If your entire online identity is anonymous, it has nothing to link your real identity to.

Everyone who uses their real name on a website has never been truly anonymous. Ever.

Shit used to be common knowledge that you do not use your real name and you do not share photos online. Now that's what 90% of people exclusively use the internet to do.

[–] ParlimentOfDoom@piefed.zip 1 points 12 hours ago

LLMs don't work like this at all. They're not analysis machines. They're probability based glorified autocorrectors.

[–] Sims@lemmy.ml 11 points 22 hours ago

Ah, 'The Guardian' accidentally wrote "Hackers" instead of "US Oligarchy" or "Corporations". Better to hide that fact by deferring the actor to "those pesky 'hackers' " - they are always anonymous.

The Guardian are totally removed information for the Oligarchs - as usual..

[–] nymnympseudonym@piefed.social 2 points 15 hours ago

I'm sorry, this paper is redundant and dumb.

We've had Stylometry tools like R-Stylo for over a decade now, as well as defensive tools against it.

[–] Randomgal@lemmy.ca 1 points 21 hours ago

Just use a VPN bro! Lmao