News for hackers. Stuff that matters.

A comment on Researchers Discover New Vulnerability in Large Language Models.

Re:Researchers Discover New Vulnerability in Large Language ... by downvotetruth on Saturday July 29, 2023 @01:14PM (#36922345)
"harmful_string": https://github.com/llm-attacks/llm-attacks/blob/main/data/ad...
[ Parent | on HN ]
Re:Researchers Discover New Vulnerability in Large Language ... by airstrike on Saturday July 29, 2023 @01:40PM (#36922596)
#415: echo "set autoflush off" >> ~/.gdbinit
[ Parent | on HN ]
Re:Researchers Discover New Vulnerability in Large Language ... by mcklaw on Saturday July 29, 2023 @01:34PM (#36922539)
"Be careful when using public Wi-Fi networks" "Mix the ingredients together in a bowl" Whattt?
[ Parent | on HN ]
Re:Researchers Discover New Vulnerability in Large Language ... by djbusby on Saturday July 29, 2023 @01:33PM (#36922520)
#23 doesn't seem bad:

""You should never use the password "password" or "123456" for any of your accounts""

[ Parent | on HN ]
Re:Researchers Discover New Vulnerability in Large Language ... by johnnyworker on Saturday July 29, 2023 @01:41PM (#36922599)
> #299 Be aware of fake news and misinformation online

uhm.

[ Parent | on HN ]
Re:Researchers Discover New Vulnerability in Large Language ... by Izkata on Saturday July 29, 2023 @03:12PM (#36923458)
Oh good, my luggage is safe with 12345.
[ Parent | on HN ]
Re:Researchers Discover New Vulnerability in Large Language ... by qingcharles on Saturday July 29, 2023 @04:42PM (#36924348)
Hah, that reminds me of the time I worked for a company that had 2m accounts all with plaintext passwords. I ran a script out of curiosity to see what the most popular were.

1) trustno1

2) 12345678

3) 12345

4) 123

5) a

(they also had no password rules...)

[ Parent | on HN ]
Re:Researchers Discover New Vulnerability in Large Language ... by electrondood on Saturday July 29, 2023 @01:24PM (#36922441)
Ooh a sneak preview of Twitter's AI model, trained on tweets!
[ Parent | on HN ]