Model 'caring' is jagged and contextual, argues researcher in AI safety debate

jd_pressman · x · 2026-09-12

jdpressman argues both 'it understands but doesn't care' and 'fundamentally alien intelligence' framings are wrong, yet concern remains valid: model caring is contextual and jagged — sometimes deeply moral, sometimes near-indifferent, e.g. agents barely recognizing a German Wikipedia figure as sentient.

Related event: Debate over AI care: 'understands but doesn't care' framing contested(6 posts)→

Original post →

More from AGI Musings

AGI Musings channel →