April 19, 2026
When bots give us the ick
The Uncanny Valley and the Rising Power of Anti-AI Sentiment
Experts say AI helps them; everyday folks say it’s creepy, risky, and wasting our time
TLDR: Surveys say experts feel helped by AI while most Americans increasingly fear it, especially in daily life and schools. Comments roast the polls’ simplicity, warn of near-term help vs long-term harm, gripe about time-wasting chatbots, and joke about an AI culture war—because trust, jobs, and control are on the line.
The vibes are off. While AI experts cheer, the public is side‑eyeing hard: in a Pew survey, 76% of experts said AI helps them personally, but only 24% of Americans agreed. A new Quinnipiac poll says 55% now think AI does more harm than good day‑to‑day (up from 44% last year), and 64% say it hurts schools. The comment section lit up like a Black Friday sale. Userbinator slammed the polls as oversimplified—“AI” is everything and nothing—so a yes/no question misses the real story. Others brought the existential dread: AI feels human until it doesn’t, and that “uncanny valley” ick is spreading from robots to everything—chatty text, too‑perfect voices, videos that crumble on close look. As one poster put it, “nobody likes to have their time wasted,” and folks are sick of friendly bots that talk big but fumble basics.
Hot takes flew. Jmathai’s mood: AI can help now and still harm later—people get hands‑on and their optimism curdles. Xnx nailed the hypocrisy meme: we want AI for our job, not for others’ jobs, and we definitely don’t want to wade through AI sludge that’s faster to make than to read. Noosphr went full culture‑war prophecy: future Thanksgiving fights about “AI equality.” Beneath the jokes is real heat: trust, jobs, and control—and a growing public disgust that’s starting to look like a movement.
Key Points
- •Pew’s 2025 survey found 76% of AI experts expected personal benefits from AI, versus 24% of the U.S. public, with the public more likely to anticipate harm.
- •A March 2026 Quinnipiac poll reported 55% of Americans believed AI would do more harm than good in daily life (up from 44% in April 2025), and 64% said harm outweighed good in education.
- •The essay argues public hostility to AI exceeds typical tech skepticism due to factors like fraud, misinformation, privacy invasion, concentration of power, and job displacement.
- •Uncanny valley theory is used to explain growing aversion: near-human AI outputs trigger human expectations that are repeatedly violated, producing mismatch and distrust.
- •Disgust, pathogen-avoidance, and broader danger-avoidance frameworks are discussed as mechanisms behind uncanny responses, with reference to a 2025 virtual agents study.