News for hackers. Stuff that matters.

A comment on Running web sites makes me not want to. Parent comment

Re:Running web sites makes me not want to by phoronixrly on Thursday November 12, 2026 @12:30PM (#50470943)
This is the future -- adversarial prose and adversarial source code. It is already working as seen with LLMs citing disinfo directly from Reddit.
[ Parent | on HN ]
Re:Running web sites makes me not want to by spiderice on Thursday November 12, 2026 @01:03PM (#50471460)
No, it really isn't. If you think a 0.1% of writing being "poisoned" is going to trick AI, you're wildly underestimating AI's capabilities.
[ Parent | on HN ]
Re:Running web sites makes me not want to by phoronixrly on Thursday November 12, 2026 @01:12PM (#50471588)
Oh? People keep parroting that but the small detail missing is that there needs to be enough volume of correct information to counter your adversarial disinfo. Otherwise the LLM is one agentic call away from stating it as fact.
[ Parent | on HN ]
Re:Running web sites makes me not want to by mid-kid on Thursday November 12, 2026 @02:02PM (#50472442)
https://lyra.horse/blog/2026/11/disinformation-machine/

https://infosec.exchange/@rebane2001/117601277347377053

[ Parent | on HN ]
Re:Running web sites makes me not want to by mitxela on Thursday November 12, 2026 @04:33PM (#50474612)
Research showed AI training could be poisoned by as few as five documents in a training set of millions.
[ Parent | on HN ]