blueteamsec

670 readers

24 users here now

For [Blue|Purple] Teams in Cyber Defence - covering discovery, detection, response, threat intelligence, malware, offensive tradecraft and tooling, deception, reverse engineering etc.

founded 2 years ago

MODERATORS

digicat

Forensic Implications of Localized AI: Artifact Analysis of Ollama, LM Studio, and llama.cpp (arxiv.org)

submitted 3 days ago by digicat to c/blueteamsec

1 comments fedilink hide all child comments

you are viewing a single comment's thread
view the rest of the comments

[–] Paragone@lemmy.world 0 points 3 days ago

They're ONLY looking-at LLM-was-used-to-commit-crime perspective??

What about LLM-committed-the-crime?!

LLM-pushed suicides are an actual thing, now..

So is LLM-pushed extremism,

so is LLM-pushed psychosis/conspiricism/delusion.

( mix that in with THC-amplified psychosis, a well-documented fact, & you've got a feedback-loop we didn't need to invest-in.

Real-world clinical scenarios underscore why this issue is so important. In one case, a young man with a long history of daily cannabis use began experimenting with “dabbing,” a method of inhaling extremely concentrated THC. Within weeks, he became intensely paranoid and hypervigilant—convinced that strangers in a black truck were following him—and grew too anxious to sleep at night. His downward spiral was halted only after he was hospitalized, treated with antipsychotic medication, and most importantly, stopped using cannabis. Once the high-THC dabs were out of his system, his paranoia and delusions quickly resolved.

from https://www.psychiatrypodcast.com/psychiatry-psychotherapy-podcast/episode-240-cannabis-thc-psychosis-risk )

_ /\ _