A growing divide between public and expert sentiment reveals that many Americans increasingly view artificial intelligence as a source of harm rather than a personal benefit.

Key Points

• A 2025 Pew survey found 76% of AI experts believe the technology benefits them, compared to only 24% of the general public.
• Quinnipiac data from March 2026 shows 55% of Americans believe AI does more harm than good, an increase from 44% in April 2025.
• Public concerns are driven by fears of job displacement, privacy invasion, misinformation, and the concentration of corporate power.
• Psychological theories, including the "uncanny valley," suggest that AI’s failure to consistently meet human social expectations triggers visceral feelings of disgust and unease.
• Experts often evaluate AI through a lens of technical utility, while the public frequently experiences the technology as an intrusive or hollow imitation of human behavior.

Why it Matters

The widening gap between expert optimism and public apprehension suggests that technical reassurances may fail to address the deep-seated emotional and existential discomfort caused by AI. If developers continue to prioritize human-like mimicry, they risk triggering persistent aversion that could hinder the long-term social and commercial adoption of these technologies.