an open source intelligence explosion is unsurvivable. any paths we have towards the good ending must avoid it at all costs.
doing it silently seems bad, but in general i am very happy anthropic is taking strong measures to prevent distillation and progress in open source models, and i hope openai will do the same.









