Jeff Ladish: researchers building superintelligence being intellectually siloed is extremely risky

JeffLadish · x · 2026-09-09

AI safety researcher Jeff Ladish argues the real issue in frontier-lab public discourse isn't how much time researchers spend on Twitter but intellectual siloing: researchers building superintelligence who lack public engagement and external scrutiny pose an extreme risk. Laneless counters that more frontier-lab tweeting may not improve the moral trajectory of the lightcone.

Related event: Researchers Debate Whether Frontier Lab Staff Should Discuss AI Publicly(2 posts)→

Original post →

More from AGI Musings

AGI Musings channel →