NBC News poll of 1,000 registered voters: just 26% had a positive view of AI, while 46% had a negative view, the third worst net negative score of all topics
Artificial intelligence has permeated nearly every part of daily American life. It's being adopted across the professional sphere …
Context & Ripple Effects
This result extends an earlier divide between public and expert expectations: a 2025 survey found more US adults expected AI to harm the country over the next 20 years, while most AI experts expected a positive effect public and expert expectations diverged on AI's long-term impact. Concern is especially salient in information work, where respondents anticipated AI would hurt news and reduce journalist employment concerns about AI's effect on news jobs.
The poll matters because AI adoption is moving into professional work and other public-facing systems while its social license remains weak. It makes public acceptance—not just technical capability or cost—a meaningful constraint on deployment.
First-order effects
- AI companies, employers, and public officials now face a clear negative-sentiment benchmark when explaining or rolling out AI-enabled products and workplace changes.
- The result strengthens the case for deployment messaging centered on concrete user benefits, safeguards, and human oversight rather than AI's novelty alone.
Second-order effects
- Organizations introducing AI into jobs, media, or essential services are likely to encounter more stakeholder scrutiny, raising the value of pilots and evidence that specific uses improve outcomes without displacing accountability.
- Local and political resistance can become a practical constraint on AI infrastructure and deployment plans when general skepticism is paired with perceived impacts on work or communities.
Third-order effects
- If negative sentiment persists as adoption broadens, AI competition will increasingly turn on trust, governance, and distribution through institutions that users already rely on—not solely model performance.
- The industry may face a widening legitimacy gap: capability gains can accelerate adoption by firms while weakening public support for the infrastructure, labor changes, and policy choices that sustain it.
The trend: AI is shifting from a technology-adoption story to a social-license test, with public trust becoming a constraint on how quickly and where it can be deployed.