Conversation

I'm seeking a researcher/journalist with an interest in AI/rationalism/xrisk for a 1-2 week project. To apply please email job@nat.org with evidence that you (a) read scifi and (b) understand this tweet:
Quote Tweet
Replying to @michaelshermer @ylecun and @ESYudkowsky
What's you response to Orthogonality Thesis+Instrumental Convergence+AI arms race+ likely Mesa Optimization+ rapid recent progress in AI via ML = we are likely doomed? I have a PhD in econ so I do understand risk.
Show replies
⟂ is real, for physics too. Biology has been more niche-filling than convergent. Awareness is discursive contextual optimization at many scales. Poor planning threatens mutual destruction: an old risk. Planners avoid ∅, but shorter cycles need fast, myriad negative feedback.
Show additional replies, including those that may contain offensive content
Show