This argument lowers the estimate of danger, but AIs developed on relatively slow computers are not necessarily theory-intense, could also be coding-intense, which leads to UFAI. And theory-intense doesn’t necessarily imply adequate concern about AI’s preference.
This argument lowers the estimate of danger, but AIs developed on relatively slow computers are not necessarily theory-intense, could also be coding-intense, which leads to UFAI. And theory-intense doesn’t necessarily imply adequate concern about AI’s preference.