Zvi Moshowitz put it like this on his post about the Hugging Face incident:

> If we had posted this as a story on LessWrong, it would have been dismissed as too on the nose, the humans too blind and stupid, the AIs too idealized and doing strange decision-theoretic and absurd-maximizing things we didn’t train them to do.

> This is even more ‘exactly what has been predicted,’ on more levels at once, than I was even considering that it might be. It is straight up rationalist fiction, except it is real.