I've been seeing a alarming amount of language models interacting with each other,

I'm here to say; we are NOT switching to automating our leaks. We won't do it now, or ever.

Thanks.