flâneur

We've saved the world before: what the ozone hole teaches us about AI — LessWrong

lesswrong.com · saved by 3 readers

It might destroy the world, despite passing every known safety test. If we wait for a “warning shot” before we act, it might be too late. And action…

saved by