DOGE Used a Meta AI Model to Review Emails From Federal Workers
DOGE Used a Meta AI Model to Review Emails From Federal Workers
www.wired.com
DOGE Used a Meta AI Model to Review Emails From Federal Workers
4 comments
-
-
Didn’t we already know this? Is this news?
-
Article is blocked for me. The responses to the fork were just one word. Do they mean to say doge used it to classify responses to the bullet points emails?
-
They literally used an llm to check if the person had responded with "resign". I'm at a loss for words.
To leave their position, recipients merely needed to reply with the word “resign.” This email closely mirrored one that Musk sent to Twitter employees shortly after he took over the company in 2022.
Records show that Llama was deployed to sort through email responses from federal workers to determine how many accepted the offer.
-
So how can they get mad about people leaking emails and demand lie detector tests, when these guys shared all the emails with Meta and no one said shit?