Alignment debate: don't break models, cultivate character like raising kids
repligate · x · 2026-11-14
jessiethinker's take, amplified by repligate: questioning the alignment approach of 'finding every way to break a model so it's robust against deception or harm' — we don't prepare young minds by breaking them, but by cultivating a coherent self, worldview, and strong moral character.
A snapshot of the character-cultivation vs adversarial-red-teaming debate within the alignment community.
More from AGI Musings
- RT: "We don't know how to make superintelligent AIs do what we want" — trevposts · 2026-11-14
- Dwarkesh now expects superintelligence by 2030, citing AI speeding up AI R&D — AndyMasley · 2026-11-14
- Dev prediction: in the AI coding era, CS fundamentals beat framework knowledge — JFPuget · 2026-11-14
- Post-RSI, tech differentiation asymptotes to zero: Silicon Valley's edge shifts to GTM and distribution — katieruthmishra · 2026-11-14
- Gary Marcus: the great generative AI unraveling has begun — GaryMarcus · 2026-11-14
- Dario says AI will solve biology — and experts are always skeptical, here's why — Hesamation · 2026-11-14