the-citizen comments on A forum for researchers to publicly discuss safety issues in advanced AI

[the-citizen]

For the record, my current position is that if MIRI doesn’t think it’s central, then it’s probably doing it wrong.