Hello,

Nvidia, SpaceX, Microsoft and a host of other tech firms have come together to launch an AI safety initiative.

This comes in the wake of a Chinese open-weight AI model defending against a recent cyberattack by OpenAI models.

AI startup Hugging Face was not able to use US frontier models to defend itself from the unprecedented hack.

Frontier systems built by Anthropic and OpenAI can only be accessed through specific infrastructure.