A new Pew Research study shows that Americans are still approaching artificial intelligence with caution, even as the technology becomes a routine part of daily life. Only 16% of respondents believe AI will have a positive effect on society over the next 20 years, while about 40% expect a negative outcome.
The survey also points to a broad trust gap. Around 67% of adults do not believe the U.S. government will regulate AI effectively, and 59% say they do not trust companies to develop it safely. Younger adults appear especially skeptical: among people under 30, just 14% expect a positive social impact from AI.
At the same time, adoption continues to rise. Roughly one in four Americans say they use AI chatbots every day, mainly for research or work. ChatGPT remains the most widely used tool, with 44% of U.S. adults reporting use. Gemini, Copilot, and Meta AI follow behind.
The report also highlights clear differences in how people engage with the technology. Men are more likely than women to use AI chatbots regularly and to express enthusiasm about them. Meanwhile, about six in ten respondents say they now read AI-generated summaries online, showing how quickly AI is shaping information habits.
Still, nearly half of Americans say they do not use AI in daily life, and older adults are far less likely to engage with chatbot tools. Among those 65 and older, nearly three-quarters say they never use them. The findings suggest that AI is expanding fast, but public confidence is evolving more slowly. That balance will likely shape how AI is integrated into everyday life in the years ahead.
