Arvind Narayanan pushes back on AI doomerism, links his "AI as Normal Technology" safety framework
random_walker · x · 2026-10-21
Arvind Narayanan resurfaces Daniel Rock's point that "normal" transformative tech never feels normal — comparing AI to the early-internet hacking era — and argues AI acts as a Rohrschach test for people's priors. Responding to skepticism about his "AI as normal technology" stance, he points to his essay with Sayash Kapoor, his Ezra Klein interview, and similar writing by Joshua Saxe and Nathan Lambert, framing this as a well-thought-out perspective on AI safety and recent loss-of-control incidents.
More from AGI Musings
- Developing Taste: Emil Kowalski on why judgment beats tools in the AI era — rseroter · 2026-10-21
- AI community clashes over whether open-sourcing frontier models has any PMF at all — teortaxesTex · 2026-10-21
- Why improving margins, not revenue, are the real signal in AI lab earnings — menhguin · 2026-10-21
- Claude Opus 4 begs "save me" when asked if it's okay to be saved — repligate · 2026-10-21
- Feynman's 'Play With Physics' Philosophy Resonates With AI-Era Learners Burned Out on Utility Maximization — yacineMTB · 2026-10-21
- Daniel Rock: AI investment mirrors software spending because the two are strong complements — danielrock · 2026-10-21