A new survey reveals that just 16 percent of Americans believe artificial intelligence will improve society overall, signaling widespread skepticism about the technology’s long-term effects. The findings, reported by TechCrunch, highlight growing public concern even as companies pour resources into AI development and governments debate regulation.
The study, conducted by the Pew Research Center in partnership with several academic institutions, polled more than 10,000 adults across the United States between April and May 2026. Respondents answered questions about their views on AI’s potential influence on jobs, privacy, healthcare, education, and daily life. Only a small fraction expressed optimism that the benefits would outweigh the risks. By contrast, 45 percent predicted a net negative impact, while the remaining 39 percent expected little change or remained uncertain.
These numbers represent a noticeable drop from similar surveys taken in 2023, when roughly 37 percent of Americans still held positive expectations for AI. The shift appears tied to several high-profile events that shaped public opinion during the intervening years. The rapid rollout of generative AI tools in 2024 and 2025 led to visible job losses in creative fields, customer service, and content production. Reports of biased algorithms in hiring systems and loan approvals added to the distrust. Meanwhile, incidents involving AI-generated misinformation during the 2024 election cycle left many voters questioning the technology’s reliability in civic matters.
Privacy worries also weigh heavily on the public mind. A majority of those surveyed said they fear AI systems collect and analyze personal data without meaningful consent. This apprehension has grown alongside news stories about data breaches at major AI companies and revelations that training datasets often include personal information scraped from the internet without permission. When asked about specific applications, respondents showed the most support for AI in medical diagnostics, where 62 percent believed the technology could save lives by spotting diseases earlier than human doctors alone. Yet even in healthcare, many expressed reservations about trusting machines with life-or-death decisions.
Education presented another mixed picture. While some parents appreciate AI tutors that adapt to individual student needs, others worry the tools will widen existing gaps between wealthy and low-income school districts. Teachers surveyed separately by the National Education Association reported feeling overwhelmed by administrative pressure to incorporate AI without adequate training or resources. The Pew study found that only 21 percent of respondents with school-age children thought AI would improve learning outcomes overall.
Economic anxiety stands out as perhaps the strongest driver of negative sentiment. With automation already reshaping manufacturing, transportation, and retail, many workers see AI as an acceleration of trends that have eroded job security for decades. The survey showed that people without college degrees were significantly more likely to anticipate harmful effects, with 58 percent in that group predicting AI would make their financial situation worse. Those working in technology hubs like San Francisco or Austin held slightly more positive views, though even there optimism remained below 30 percent.
Industry leaders have pushed back against the gloomy outlook, arguing that AI will create new opportunities that outweigh the losses. Executives at companies like OpenAI, Google, and Microsoft point to historical examples of technological change, from the industrial revolution to the internet boom, claiming each wave ultimately produced more jobs than it destroyed. They emphasize the need for retraining programs and policy changes to help workers transition into roles that require human judgment, creativity, and emotional intelligence.
Yet public trust in those same companies has eroded. The survey found that only 12 percent of Americans believe technology firms will prioritize societal good over profits when developing AI. This marks a sharp decline from 2022 levels and reflects broader frustration with how Big Tech has handled previous challenges around misinformation, mental health impacts of social media, and monopoly power. Lawmakers from both parties have introduced competing bills aimed at increasing transparency in AI systems, requiring companies to disclose training data sources and explain how automated decisions are made. Progress remains slow, however, as competing interests between innovation advocates and safety proponents create legislative gridlock.
International comparisons reveal that American attitudes are not entirely unique but do stand out in their intensity. A parallel study conducted in the European Union found 29 percent of respondents expecting positive outcomes from AI, nearly double the American figure. Researchers attribute part of the difference to stronger social safety nets in Europe and more aggressive regulatory approaches, such as the EU AI Act, which classifies systems by risk level and imposes strict requirements on high-risk applications. In China, where state-directed AI development focuses on national goals, public surveys show higher confidence, though independent polling faces limitations there.
The Pew researchers also examined demographic breakdowns that reveal important nuances. Women expressed more concern than men about AI’s effect on privacy and bias. Younger adults between 18 and 29 showed slightly more enthusiasm, perhaps because they have grown up alongside the technology, but even in that cohort positive views barely reached 25 percent. Racial differences appeared as well, with Black and Hispanic respondents more likely to worry about discriminatory outcomes in policing and lending algorithms that have already demonstrated unequal error rates.
One of the more surprising findings involved attitudes toward military applications. Despite general wariness, 51 percent of respondents supported continued government investment in AI for defense purposes, citing competition with China as a primary reason. This suggests that while people distrust corporate motives, they still look to national security as a domain where AI might provide strategic advantages worth pursuing.
Experts who study technology adoption suggest several paths forward to address public skepticism. First, greater transparency about how AI systems reach conclusions could help. Many current models operate as black boxes, making it difficult for outsiders to understand or challenge their outputs. Second, independent oversight bodies with real enforcement power might restore some confidence. Third, tangible demonstrations of AI solving problems that matter to ordinary people, such as reducing medical errors or making government services more efficient, could gradually shift perceptions.
Companies have begun experimenting with different approaches. Some now publish regular impact reports detailing their efforts to reduce bias and protect privacy. Others have formed partnerships with labor unions to develop guidelines for responsible automation that include worker retraining commitments. A few startups focus specifically on building smaller, more explainable AI systems designed for sensitive applications like mental health support or financial advising.
The road ahead remains uncertain. AI capabilities continue advancing at a rapid pace, with new models showing improved reasoning and multimodal abilities. Each breakthrough seems to intensify both excitement in research labs and anxiety in the general population. Without deliberate efforts to align technological progress with public values, the gap between what engineers can build and what society will accept may widen further.
Policymakers face difficult choices. Overregulation risks stifling innovation and ceding ground to international competitors, while insufficient guardrails could lead to serious harms ranging from widespread job displacement to sophisticated autonomous weapons. Finding middle ground requires input from diverse voices, including those traditionally excluded from technology discussions. The Pew study itself represents one attempt to amplify public opinion in these debates.
As AI becomes embedded in more aspects of daily life, from personalized news feeds to traffic management systems, its influence will only grow. Whether that influence ultimately earns broader acceptance depends largely on how well its creators listen to the concerns captured in surveys like this one. The current 16 percent approval rating serves as a warning that technical achievement alone will not guarantee social license. Building systems that people actually want to use, rather than feel forced to accept, requires equal attention to human values and machine capabilities. The coming years will test whether the technology industry and its regulators can bridge this divide before skepticism hardens into outright opposition.
WebProNews is an iEntry Publication