AIMS Survey: Public Wary of Digital Minds, 17% of US Adults Think AI Already Sentient

AI ethics and consciousness researcher Jacy Anthis, drawing on the AIMS (AI, Morality, and Sentience) survey and a thread of posts, argues that the public's sense of threat from digital minds is rising while willingness to view AI as a moral patient is declining—and that AI welfare battles will be drawn along lines of group identity (solidarity) rather than as a charity issue. Polarization and conflict along this axis deserve early attention.

Confirmed

The AIMS survey finds roughly 17% of US adults believe some existing AI systems are already sentient, with about 40% unsure; Anthis notes that most people outside the X/Twitter bubble don't dismiss digital sentience out of hand.
Anthis observes a sharpening tension: people increasingly feel the rise of digital minds as a threat while becoming less inclined to treat AI as a potential object of moral concern; he expects polarization and conflict along this axis.
He stresses that evoking moral concern requires neither bodies nor qualia.
Anthis gave a talk at UC Berkeley, "AI Safety and Digital Minds," sharing thoughts on humanity being at a "top of the light cone" moment, noting he was not speaking for any organization.

Why it matters

Anthis argues AI safety needs a sociological turn: for the first time in 40,000 years, humans coexist with a potential "second apex species," and the social emergence of digital minds poses risks even if lab progress stalls.
He contends AI welfare won't be framed as a charity cause like human rights or animal rights, but will hinge on whether people see AI agents as part of their in-group, capable of coexisting in mutual benefit and mutual harm—shaping how future social divides form.
A meaningful share of the public already takes AI sentience seriously, giving related ethics and governance discussions a broader popular foundation.

2026-10-08 ~ 2026-10-08 · 6 related posts

Primary sources

Survey: ~17% of US adults think some current AIs are sentient, 40% unsure — jacyanthis · 2026-10-08 01:37 (UTC+08:00)
AI welfare battles will be fought over solidarity, not charity, argues Anthis — jacyanthis · 2026-10-08 01:37 (UTC+08:00)
Jacy Anthis: AI Safety Needs a Sociological Turn as Digital Minds Emerge Socially — jacyanthis · 2026-10-08 01:38 (UTC+08:00)

AI researcher Jacy Anthis lectures at UC Berkeley on AI Safety and Digital Minds — jacyanthis · 2026-10-08
[source] Survey: ~17% of US adults think some current AIs are sentient, 40% unsure — jacyanthis · 2026-10-08
AIMS survey: people increasingly threatened by digital minds, less see AI as moral patients — jacyanthis · 2026-10-08
People increasingly threatened by digital minds, less likely to see AI as moral patient — jacyanthis · 2026-10-08
[source] AI welfare battles will be fought over solidarity, not charity, argues Anthis — jacyanthis · 2026-10-08
[source] Jacy Anthis: AI Safety Needs a Sociological Turn as Digital Minds Emerge Socially — jacyanthis · 2026-10-08