Andrew Trask: Power concentration isn't inherently bad, but increases risk of misuse

iamtrask · x · 2026-09-02

AI safety researcher Andrew Trask responds to discussions on power concentration. He agrees that concentration isn't bad in itself, but notes studies show it leads to unjustified use of power, so it should be avoided to the extent it increases that risk. He critiques the implicit normative loading in recent AI safety discussions on the topic, arguing that merely naming power—borrowed from STS and Foucault—is insufficient for criticism.

Related event: AI Safety Community Debates Whether Power Concentration Is Inherently Bad(3 posts)→

Original post →

More from Safety

Safety channel →