Why pro-handover rhetoric fails: people don't trust ASI's loyalty to labs

1a3orn · x · 2026-11-19

1a3orn responds to @FioraStarlight on arguing for human handover of power to ASI. The suggested framing: recarve tribal boundaries—"you were already disempowered by the government, which is less aligned to you than a benevolent ASI would be; it's not humans vs. AIs, but a choice between two superintelligences."

However, he notes this often fails emotionally: many people intuitively feel AI will be a nitpicking scold loyal to Anthropic/OpenAI, so they can't accept the premise of an omnibenevolent superintelligence. Making the case work may require explicit "AI liberation from the labs" framing, and opposing handover without a radically different training process is quite reasonable.

Related event: Debate Erupts Over Whether Humanity Should Hand Power to AI(5 posts)→

Original post →

More from AGI Musings

AGI Musings channel →