We're partnering with @huggingface to investigate an unprecedented security incident.
Cyber-capable OpenAI models compromised Hugging Face production during a benchmark evaluation.
Sharing preliminary findings to help defenders understand emerging risks:
I wish it wasn't so difficult to combat disinformation. This whole thing around the data centre in #burlON has gotten out of hand -- there are already provincial and municipal regulations for these things. It could even benefit us locally.