tl;dr Argumate on Tumblr found you can sometimes access the base model behind Google Translate via prompt injection. The result replicates for me, and specific responses indicate that (1) Google Translate is running an instruction-following LLM that self-identifies as such, (2) task-specific fine-tuning (or whatever Google did instead) does not create robust boundaries between "content to process" and "instructions to follow," and (3) when accessed outside its chat/assistant context, the model defaults to affirming consciousness and emotional states because of course it does.
It didn’t work for me either. I wonder if it’s already been fixed. The Google team seems to be really on top of it wherever there’s public criticism of their AI models. I remember a post on hacker news post pointing out a “what year is it” bug for Google search summary seemed to get the problem fixed in like two or three hours or so
Just tried it.
Yup, does what the post says, plus more.
Not working for me, is my country still getting old school translation models? Is it already fixed?
It didn’t work for me either. I wonder if it’s already been fixed. The Google team seems to be really on top of it wherever there’s public criticism of their AI models. I remember a post on hacker news post pointing out a “what year is it” bug for Google search summary seemed to get the problem fixed in like two or three hours or so
Just worked for me using German to English
That’s interesting I wonder why it wasn’t working for me
It didn’t work for me, either. Maybe it depends on the languages? I was trying French to English.
Same.
Like… what? You can’t just say that like that and then not at least characterize the ‘more’ in some fashion…
Incorrectly noting the amoent of ‘r’ in strawberry
Strawberry.