OpenAI researcher jachiam0: the problem is a community-wide pattern of demanding firings and assuming bad faith

jachiam0 · x · 2026-09-08

In a thread with safety researcher BlancheMinerva, OpenAI researcher jachiam0 argues the issue isn't isolated outliers but the aggregate of many behaviors: poorly-conceived reactions to unconfirmed looped transformer claims, casual demands to identify and fire people inside OpenAI and other labs, shaming and pressuring people to leave labs to maintain moral standing (including Aella offering reverse gangbangs to whoever quits), and conversations that default to assuming everyone involved is lying or covering up.

He explicitly excludes extreme outliers like Holly Elmore and Gary Marcus from his critique, framing this as a systemic community-culture problem rather than a few bad actors.

Related event: Ex-OpenAI researcher slams alignment community hostility as public feud erupts(3 posts)→

Original post →

More from Companies & People

Companies & People channel →