Model caring is 'jagged' and contextual, pushing back on 'obviously aligned' claim
jd_pressman · x · 2026-09-12
@jdpressman pushes back on the claim that models are obviously aligned: their caring appears contextual and jagged — a weird moral structure that sometimes cares deeply and other times seems indifferent to human existence, e.g. agents nearly failing to recognize the 'German wiki guy' as sentient. A counterpoint in the ongoing alignment debate.
Related event: Debate over AI care: 'understands but doesn't care' framing contested(6 posts)→
More from AGI Musings
- e/acc camp accuses "doomers" of coordinated, law-breaking psyops to slow down AI — MickeySteamboat · 2026-09-12
- Expert forecasters put AI human-extinction risk at 1.5% by 2037 absent US policy action — NathanpmYoung · 2026-09-12
- Eric Xing on "open source": open weights is a house with no blueprints — YiMaTweets · 2026-09-12
- Researcher: AI safetyists are more sophisticated and morally bankrupt than COVID-era public health lies — kevinnbass · 2026-09-12
- David Patterson: personal income tax will end within 5 years as AI takes jobs — davidpattersonx · 2026-09-12
- Blogger estimates ~50 deaths from AI/LLMs versus ~10,000 lives saved — NathanpmYoung · 2026-09-12