←back to thread

Anthropic’s paper smells like bullshit

(djnn.sh)
1160 points vxvxvx | 1 comments | 16 Nov 25 11:32 UTC | HN request time: 0.281s | source

Earlier thread: Disrupting the first reported AI-orchestrated cyber espionage campaign - https://news.ycombinator.com/item?id=45918638 - Nov 2025 (281 comments)
Show context
neuroelectron ◴[16 Nov 25 12:26 UTC] No.45944599[source]▶
So Claude will reject 9 out of 10 prompts I give it and lecture me about safety, but somehow it was used for something genuinely malicious?

Someone make this make sense.

replies(4): >>45944607 #>>45944631 #>>45944677 #>>45944745 #
cbg0 ◴[16 Nov 25 12:52 UTC] No.45944745[source]▶
I've never had a prompt rejected by Claude. What kind of prompts are you sending where "9 out of 10" get rejected?
replies(2): >>45945433 #>>45950230 #
1. yahoozoo ◴[17 Nov 25 01:57 UTC] No.45950230[source]▶
If you ask it to help you write a bot/cheat for a video game it will usually refuse due to breaking the games terms of service, etc.