Elon has publicly stated that he cares a great deal about safety. He has stated that the only safe models are those which align greatest with truth, that which is in reality. In this, xAI has lived up, as it has proved to hallucinate least (or close to least) in benchmarks.
replyIf you read that, quote again, he is saying "how can you quantify safety in a card?"
Seriously though, why is it a model "card", safety "card"? I had to lookup to learn that it comes from HuggingFace's vague definition of "README" in the model's repo. This is such a specific thing that I don't think anyone except a very small population would know - not the users, not the c-suites.
replyI don't like Musk or Grok. But not knowing what's a safety card is not a signal of anything IMO.
He asked why it would be a card. URL slug of world’s hottest (non-Nvidia?) company:
reply system-cards
You’d have to be asleep at the wheel. For years:
Claude 2
July 2023
Read system card