A new survey from Anthropic reveals a striking paradox: Americans see enormous promise in AI for curing diseases and helping people with disabilities, yet they harbor deep anxieties about job displacement and corporate accountability. The company's first Anthropic Public Record survey, conducted in November and December of 2025 with nearly 52,000 Americans, offers a detailed snapshot of how the general public views artificial intelligence as the technology becomes increasingly woven into daily life.

What Are Americans' Biggest Hopes and Fears About AI?

When asked to identify their top three hopes for AI, nearly half of Americans, 48%, ranked curing diseases like cancer or Alzheimer's as a leading priority. This was followed by helping people with disabilities at 36%, and then technological progress and making life easier in general, each cited by 23% of respondents.

But hope and fear exist in tension. The survey uncovered a starkly different picture when researchers asked about concerns. Job loss emerged as the dominant worry, with 64% of Americans expressing concern about AI-induced unemployment. This fear proved remarkably consistent across demographic lines, appearing as the top concern among Democrats at 67% and Republicans at 62%, in households with and without children, and across every state from Iowa at the high end with 71% to Mississippi at the low end with 57%.

Beyond job displacement, Americans identified several other significant concerns:

• Cognitive Dependency: 56% of respondents worried that AI integration could leave people unable to think for themselves, making them reliant on artificial systems for basic reasoning.
• Misinformation: 52% expressed concern about AI's potential to spread false information at scale, amplifying existing problems with social media.
• Criminal Use: A substantial portion feared AI could be weaponized for illegal activities, from fraud to hacking.
• Surveillance: Americans worried about AI enabling invasive monitoring and loss of privacy.

Notably, Americans showed greater concern about AI misuse than about the theoretical risk of AI systems becoming misaligned with human values and "going rogue." The fears that ranked highest tended to be concrete and near-term, with historical precedent in earlier technologies like automation, smartphones, and social media.

Does Experience With AI Change People's Minds?

One of the survey's most revealing findings involves the relationship between hands-on AI experience and job loss anxiety. People who use AI tools at work every day showed notably lower concern about job displacement, with 54% worried compared to 70% of those who don't use AI at all.

This gap suggests that direct experience with AI may help people develop practical skills and fluency that allow them to augment rather than replace their work. Hands-on use may also reveal AI's real limitations, making the technology seem less like an existential threat to employment. The pattern likely reflects a combination of these factors, though the survey data points to experience as a significant moderator of fear.

Interestingly, job loss concerns actually increase with education level. Americans with postgraduate degrees were nearly 10 percentage points more worried about displacement than those with a high school education or less. This counterintuitive finding aligns with Anthropic's broader economic research, suggesting that workers whose jobs most closely overlap with AI capabilities are understandably more anxious about the technology.

How Do Americans View AI's Workplace Capabilities?

The survey presented respondents with 14 common workplace tasks and asked two questions: how well they thought AI could perform each task today, and how much AI involvement they would prefer in their own jobs. The results revealed a consistent pattern: Americans rated AI's capabilities fairly high, but remained hesitant about actual workplace deployment.

At the high end, 75% of Americans said AI was as good as or better than humans at research tasks. At the low end, 44% believed AI matched human performance in service and support roles. Yet even on tasks where AI was perceived as most capable, such as research and data analysis, nearly half of respondents said they wanted no AI involvement in their own work.

This gap between perceived capability and acceptance suggests that Americans distinguish between AI's technical competence and their comfort with its deployment. Acceptance appears to move in lockstep with perceived capability, however; the more competent AI is judged to be in a given domain, the more willing people become to use it.

What Do Americans Want From Government and Companies?

Perhaps the most striking finding concerns trust and accountability. Only 15% of Americans said they trust AI companies to make decisions about how AI is developed and used. This low confidence in corporate stewardship translates into strong support for government intervention.

Over 70% of survey respondents believe the government should play a role in regulating AI, and this support was bipartisan. When asked which areas should receive regulatory attention, Americans prioritized:

• Privacy Protection: 56% wanted government action to safeguard personal data and prevent invasive surveillance.
• Child Safety: 52% emphasized the need for protections specifically designed for minors.
• Liability for Harm: 49% called for clear legal responsibility when AI systems cause damage.

When asked what would best ensure AI benefits humanity, Americans ranked two actions highest: holding AI companies legally liable for harm at 47%, and prioritizing safety over growth at 44%. These responses indicate that the public views corporate accountability and cautious development as essential safeguards.

How to Understand Public Opinion on AI Development

For policymakers, business leaders, and AI researchers seeking to navigate public sentiment, the Anthropic Public Record survey offers several practical insights:

• Seek Broad Consensus: On most questions, AI did not heavily divide Americans along typical partisan, geographic, or educational lines. Where disagreement existed, it was largely in the intensity of people's views rather than fundamental opposition, suggesting room for common ground on AI governance.
• Address Concrete Fears First: Americans prioritize near-term, tangible concerns like job loss and misinformation over abstract risks. Policies and corporate practices should focus on these immediate harms before tackling longer-term theoretical scenarios.
• Build Trust Through Transparency: With only 15% trusting AI companies, transparency about capabilities, limitations, and safety measures could help shift public perception. Companies that openly acknowledge risks and demonstrate accountability may gain credibility.
• Engage Users Directly: The gap between AI users and non-users suggests that hands-on experience reduces anxiety. Expanding access to AI tools and education could help more people develop practical fluency and realistic expectations.

Anthropic plans to repeat the Anthropic Public Record survey regularly, evolving its scope as new topics become salient and allowing researchers to track how public attitudes shift as AI capabilities advance and adoption deepens. The company also intends to expand the survey outside the United States in the future, offering a window into how global attitudes toward AI differ from American perspectives.

The survey builds on other research underway at Anthropic, including a global qualitative study of 81,000 Claude users conducted through Anthropic Interviewer, the company's tool for conducting in-depth interviews at scale. Anthropic also regularly releases data from the Anthropic Economic Index, which draws on anonymized Claude usage data to show how people around the world are employing AI in their work and daily lives.

For now, the data paints a picture of a public eager to realize AI's promised benefits but wary of the disruption it may bring. Americans want accountability from the companies building these systems, and they expect government to step in where corporate self-regulation falls short. As AI continues to advance, bridging the gap between public expectations and corporate practice may prove essential to maintaining public trust in the technology.

" }