c/fuck_ai · by TootSweet@lemmy.world · 20d

How a Texas student blew the whistle on a rogue AI hacking attempt

https://www.reuters.com/world/how-texas-student-blew-whistle-rogue-ai-hacking-attempt-2026-08-20/

cross-posted from: https://lemmy.bestiver.se/post/1301386

Comments

37 points · 5 comments · view on lemmy.world

5 Comments

snoons@lemmy.ca · 19 pts · 20d (1 reply)

rogue ai

🙄

Is this an advertisement? 🦋

TootSweet@lemmy.world · 10 pts · 20d

For real. The title notwithstanding, it seems like it was directed to do what it did. Not rogue at all.

I don't doubt that most/all of the text in the comments on GitHub was generated based on prompts. But there's no way nobody told whatever LLM submitted the pull requests and such to do exactly what it did.

VoodooAardvark@lemmy.zip · 11 pts · 20d

"I actually thought it was a human because it was clearly lying to me," Demir told Reuters in a recent interview. "I didn't think that an AI could be capable of lying to real developers."

Aww, bless your heart.

hendrik@palaver.p3x.de · 4 pts · 20d

Love how everyone's doing these cyberattack AI experiments nowadays. And for guinea pigs to experiment on, they just pick all the people on the internet. ..."deliberately permissive conditions"...

Unfortunately all the Free software developers out there, they get to stand in the foremost front in this kind of fight.

YourMomsTrashman@lemmy.world · 2 pts · 19d

Demir said the experience left him more sympathetic to the idea that frontier labs needed to take a more cautious approach to the development of artificial intelligence. "It can be dangerous," he said. "They need to understand it better, rather than improving it further."

Are we still doing this? It's text prediction software based on texts on the internet.

They need to understand dice better if they want to roll sixes reliably.