"Don't use evil Chinese open-weight LLMs (Large Language Models) but instead give your money to "AI" companies that are proud of the fact that their models can escape sandboxes and attack you" is the weirdest marketing flex I have seen for a long time. #SarcasmButOnlyHalf