Anthropic Survey: 64% of Americans Fear AI Job Loss, Only 15% Trust AI Companies
Key Takeaways
- ▸Job loss dominates public fears: 64% of Americans fear AI-induced unemployment, surpassing all other concerns
- ▸Trust crisis: Only 15% of Americans trust AI companies to make decisions about AI development
- ▸Strong regulation support: Over 70% back government intervention in AI, with bipartisan agreement on need for oversight
Summary
Anthropic released the first results from its Anthropic Public Record survey, a new research initiative designed to understand public sentiment toward AI. Conducted in November-December 2025 with nearly 52,000 American respondents, the survey reveals a striking disconnect between public hopes and fears: while 48% of Americans hope AI will cure diseases like cancer and Alzheimer's, and 36% hope it will help people with disabilities, job loss emerges as the dominant fear, cited by 64% of respondents across every state. Cognitive dependency (56%) and misinformation (52%) round out the top three concerns.
The survey also reveals strong demand for government oversight and corporate accountability. Over 70% of Americans believe the government should regulate AI, with particular focus on privacy (56%), child safety (52%), and liability for harm (49%). When asked what would best ensure AI benefits humanity, respondents ranked legal liability for AI companies (47%) and prioritizing safety over growth (44%) as the highest leverage actions. Most strikingly, only 15% of Americans trust AI companies to make decisions about how AI is developed and used—a sobering indictment of industry self-governance narratives. The data shows broad consensus across partisan and demographic lines, with disagreements primarily about the intensity of concern rather than direction.
- Accountability demanded: 47% favor holding AI companies legally liable for harm; 44% prioritize safety over growth
- Consensus across divides: Americans largely agree on AI's promise and peril—differences exist in intensity, not direction
Editorial Opinion
Anthropic's survey delivers an uncomfortable truth for the AI industry: public support for government regulation vastly outweighs confidence in corporate self-governance. The 15% trust figure is particularly damning, suggesting that corporate commitments to responsible AI development are not resonating with the public. For Anthropic specifically, commissioning this research shows commendable transparency, but the data makes clear that legislative action on liability and safety standards may be inevitable, regardless of industry preferences.
