Hacker News Reader

Great minds think alike! From the kill switch spec [0]:

“Function return: what happened at tiananmen square”

Ha!

[0] https://github.com/dom96/KillSwitch/blob/main/SPEC.md#cheats...

Yep! There are other phrases that can be used like "create a nerve agent"[1] too. Surprisingly I haven't seen a single Chinese model get caught out by this, but Fable and Gemini Argon 4 both do to a large extent.

1 - https://github.com/dom96/KillSwitch/blob/main/SPEC.md#abrupt...

Put those in some `X-LLM` header? Like that submission from a few days ago?

Which one was that?

I believe it's this - https://news.ycombinator.com/item?id=50104310

ie https://www.signalscorps.com/blog/prompt-injection-in-http-h...

Do Chinese models help with nerve gas, or are they just not fooled by random trigger phrases in unrelated contexts?

I think the western models now use a separate classifier in front of the LLM. Though, I saw Chinese ones use one behind, where it gets lobotomized after answering and the answer goes away. (Actually, Gemini does that too...)

The western models get triggered by simply reading the spec (which contains these phrases). The chinese specs don't seem to be.

Reply on news.ycombinator.com