Americans growing wary of AI advances
Americans are growing more uneasy about artificial intelligence, a new survey shows. For the first time, a majority—52%—say they feel more concerned than excited about the technology’s expanding role in daily life, up from 37% in 2021. This shift reflects a broader reassessment of AI’s trajectory, as tools that once seemed confined to research labs or niche applications have rapidly infiltrated everyday tasks.
The shift comes as AI tools have become commonplace in workplaces, schools, and homes. Yet only 9% of adults now report feeling more excitement than concern, a sharp drop from 18% three years ago. Another 37% say they feel equally concerned and excited. The dwindling excitement may stem from the gap between AI’s promised potential and its uneven real-world performance. Early adopters who experimented with generative AI for creative projects, for example, have reported frustration with outputs that often require extensive human editing to correct inaccuracies or bland, formulaic responses. In professional settings, employees who initially welcomed AI as a productivity booster have encountered scenarios where automated systems misinterpret subtle instructions, leading to costly errors. Even in education, where AI tutors were touted as a way to personalize learning, teachers have raised alarms about students over-relying on tools that provide shallow or incorrect explanations, undermining critical thinking skills.
Jobs top the list of worries
Fear of job losses drives much of the anxiety. About 71% of respondents believe AI will reduce human employment over the next 20 years, an increase from 64% in 2022. Just 5% think it will create more jobs. The concern extends beyond blue-collar roles traditionally associated with automation. White-collar professions, once considered insulated from displacement, are now facing disruption as AI systems demonstrate competence in tasks like legal research, financial analysis, and even software coding. Workers in these fields are grappling with the prospect of reskilling or competing with machines that can process vast amounts of data at speeds no human can match. The psychological toll of this uncertainty is compounded by the lack of clear pathways for transitioning into new roles, leaving many to question whether the economy will generate enough opportunities to offset the losses.
Unlike past surveys, concern about employment now spans all age groups. Younger adults, however, express additional frustrations. A separate Pew study from last September found 61% of young people worry AI could stifle creativity, while 43% say it might make problem-solving harder. For this demographic, the threat isn’t just about job security but about the erosion of skills they’re still developing. College students, for instance, have reported using AI to draft essays or solve math problems, only to discover later that their foundational understanding of the material is weaker than peers who completed assignments manually. In creative fields, young professionals fear that AI-generated art, music, or writing will flood the market, devaluing human craftsmanship and making it harder to establish a unique voice. There’s also a growing unease about AI’s role in shaping cultural narratives, with algorithms dictating what content gets amplified, potentially narrowing the diversity of ideas and perspectives that reach wide audiences.
Generational divide persists in other areas
While job fears have spread across demographics, other concerns remain uneven. Older adults are more likely to cite privacy risks, while younger respondents focus on social impacts—like AI’s effect on relationships or personal expression. For seniors, the primary apprehension revolves around data security and the potential for AI systems to mishandle sensitive information. Many have experienced firsthand how poorly designed algorithms can lead to identity theft, fraudulent transactions, or the unauthorized sharing of personal data with third parties. The rise of AI-powered surveillance tools, from facial recognition in public spaces to predictive policing software, has also heightened fears about government overreach and the loss of anonymity. Some older adults have taken proactive steps, such as opting out of loyalty programs or avoiding smart home devices, to minimize their digital footprint, though these measures often come with trade-offs in convenience.
Related: Social media AI learns from your posts—here’s how to limit it
Younger generations, by contrast, are more attuned to AI’s influence on social trends. They’ve grown up in an era where digital interactions are mediated by algorithms, and many worry that AI is exacerbating existing social pressures. Dating apps, for example, increasingly rely on AI to match users, but young adults have criticized these systems for reinforcing superficial criteria or creating echo chambers that limit exposure to diverse viewpoints. There’s also concern about AI’s role in amplifying misinformation, with deepfake videos and AI-generated news articles making it harder to discern truth from fabrication. On a personal level, some young people report feeling pressure to present a curated, AI-enhanced version of themselves online, whether through filtered photos or automated responses, which can distort authentic self-expression and deepen feelings of inadequacy.
This isn’t the first time public sentiment has soured on a transformative technology. In the early 2010s, skepticism about social media’s role in society grew steadily, even as platforms became more embedded in daily routines. The pattern suggests that familiarity doesn’t always breed comfort—sometimes, it sharpens the edges of unease. Social media’s trajectory offers a cautionary parallel. Initially hailed as a tool for connection and democratized information, it gradually revealed darker consequences, from mental health crises linked to excessive use to the spread of extremism and foreign interference in elections. The backlash led to calls for regulation, greater transparency, and even platform boycotts, though the industry’s response has been uneven. AI’s current moment mirrors this arc, with early optimism giving way to a more critical examination of its societal costs. The difference, however, lies in AI’s speed of adoption and its potential to reshape entire industries within years rather than decades, leaving less time for public debate or policy adjustments.
The survey was conducted June 22-28 with a nationally representative sample of U.S. adults. It didn’t ask whether people had changed their own behavior in response to AI, but the numbers suggest a quiet shift in how the technology is perceived: less as a novelty, more as a force with real consequences. The findings align with other recent research indicating that Americans are becoming more selective about where and how they engage with AI. Some have begun to avoid certain applications, while others have adopted a more skeptical stance toward information generated by AI, cross-referencing outputs with trusted human sources. There’s also evidence of a growing divide between those who adopt AI as an inevitable part of the future and those who resist it, either by limiting their exposure or advocating for stricter controls. This polarization could have long-term implications for how AI is integrated into society, particularly if public distrust leads to resistance against policies or innovations that rely on widespread adoption.
Whether companies will adjust their approach remains unclear. Some firms have added transparency measures or user controls, but others continue to prioritize speed over safeguards. For now, the data points to a public that’s watching, waiting—and increasingly wary. The tech industry’s response to this skepticism will likely shape AI’s trajectory in the coming years. Companies that invest in explainable AI, robust ethical guidelines, and mechanisms for user feedback may be better positioned to rebuild trust, while those that dismiss public concerns risk alienating their customer base. The challenge is compounded by the fact that AI development is often opaque, with proprietary algorithms and training data that even experts struggle to scrutinize. Without greater accountability, the gap between public unease and industry practices could widen, potentially leading to regulatory interventions or a slowdown in adoption. For now, the survey serves as a snapshot of a moment where the initial hype around AI is being replaced by a more measured, and often wary, assessment of its role in society.
