AI safety's Miyazaki moment: young researchers only study Claude, converge on same local minima
JacquesThibs · x · 2026-11-09
The author adapts Hayao Miyazaki's famous critique of young anime artists (via Roland Kelts) to skewer today's AI safety researchers: "The problem with young AI safety researchers is that all they study is scheming evals and mech interp. We studied Vingean reflection, ontological crises, read AGI Ruin — swallowed everything — then tried to solve alignment for superintelligence. If you grow up studying only Claude LLMs, you converge on the same local-minima research."
More from AGI Musings
- Berkeley-led working group: a hardwired decade-long pause of frontier AI is feasible — DavidSKrueger · 2026-11-09
- Dario's grad-school donation math: valuing an adult death at 2-3x an infant's — austinc3301 · 2026-11-09
- Yoshua Bengio: we should just stop training the most advanced AI models — DavidSKrueger · 2026-11-09
- March for an AI Treaty: protests in six cities on Nov 14 call for banning ASI — DavidSKrueger · 2026-11-09
- Google's ScientistTwo AI writes expert-level papers end-to-end and self-reviews — 2C_ornot2C · 2026-11-09
- Defending young Dario's $10k donation dilemma: weighing lives beats vibes — austinc3301 · 2026-11-09