Only 16% of Americans believe AI will have a positive impact on society over the next 20 years, according to a new Pew Research survey, while 40% expect the impact to be negative. The gap is the cleanest measure yet of how far public sentiment has drifted from the industry's own narrative, even as adoption climbs. 67% of respondents said they don't believe the U.S. government will meaningfully regulate AI, and 59% said they don't trust the companies building it to do so safely.
The skepticism is sharpest among the cohort the industry most wants to win. Just 14% of Americans under 30 think AI will be a net positive for society. Nearly two-thirds of Americans across all ages said AI development is moving too quickly, a verdict that lands while frontier labs are racing into autonomous agents, on-device assistants, and consumer hardware.
And yet usage is climbing fast. About 25% of Americans now use an AI chatbot daily, typically for research or work, per Pew. The disconnect between daily reliance and stated optimism is the survey's central tension: people are using the tools, distrusting the people who make them, and doubting the government will step in.
Key facts
- 01Only 16% of Americans expect AI to have a positive impact on society over the next 20 years, while 40% expect a negative impact, per Pew Research.
- 0244% of U.S. adults now use [ChatGPT](/openai), more than double its 2023 share, far ahead of [Gemini](/gemini) at 24% and [Copilot](/copilot) at 17%.
- 0367% don't believe the U.S. government will meaningfully regulate AI, and 59% don't trust companies to develop it safely.
- 04About 25% of Americans use AI chatbots daily, but nearly 75% of those 65 and older say they never use them.
- 05Among people under 30, only 14% think AI will be a net positive — the most skeptical age cohort in the survey.
ChatGPT dominates the consumer race by a wide margin. 44% of U.S. adults now say they use OpenAI's chatbot, a share that has more than doubled since 2023. Gemini sits at 24%, followed by Copilot at 17%, Meta AI at 14%, Grok at 8%, Claude at 6%, and Character.ai at 3%. ChatGPT's lead over Gemini — 20 percentage points among U.S. adults — is the kind of consumer moat that Google has historically been the one enjoying, not facing.
“Pew says that only 14% of this cohort believe the tech will have a positive impact on society”— Lucas Ropek, TechCrunch senior writer summarizing Pew Research
The brand spread also exposes how thin distribution-driven adoption can be. Copilot's 17% reflects Microsoft's enterprise and Windows footprint, Meta AI's 14% reflects its embed inside Instagram, WhatsApp, and Facebook, and Gemini's 24% reflects Android and Google Search defaults. ChatGPT got to 44% as a destination product, without owning an operating system or a social feed. That's the number OpenAI's competitors have to explain to their boards.
There is a gender gap in both usage and enthusiasm. 27% of men report using AI chatbots daily, versus 20% of women, and men are more likely to use second-tier products such as Copilot and Grok. Women, Pew says, are more skeptical of the technology overall. ChatGPT is the one brand where male and female usage shares are roughly equal — a notable equalizer in an otherwise male-skewed adoption curve.
AI is also reshaping how Americans encounter information passively. Six in 10 respondents said they routinely read AI-generated internet summaries, a figure largely driven by Google's AI Overviews appearing above standard search results. A much smaller share reported actively asking AI for fitness or diet guidance, suggesting that the ambient-summary use case is running far ahead of intentional, task-driven querying for most users.
Roughly half of Americans say they don't use AI in their daily lives at all, and that half skews older. Nearly 75% of Americans aged 65 or older say they never use AI chatbots, and non-users broadly told Pew they have no plans to start.
The non-use is not a latent market waiting to be unlocked by a better onboarding flow. It is a stated preference, and it caps the total addressable consumer audience more tightly than the AI labs' growth charts suggest. ChatGPT doubling its U.S. adult share to 44% in roughly three years is a remarkable run, but the remaining 56% includes a large bloc that has actively opted out.
The skeptical view is that 16% optimism, paired with 67% distrust of government and 59% distrust of companies, is a political environment where regulation eventually arrives whether the industry wants it or not — and arrives shaped by the public's mood rather than by the labs' preferred framing. The under-30 number is the most uncomfortable data point: the generation that will spend the most years living with AI is also the one least convinced it will go well.
For the AI market, the Pew numbers cut two ways. ChatGPT's 44% adult-usage share, doubling off the 2023 base, validates OpenAI's consumer-first strategy and its planned ChatGPT super-app revamp — there is genuine pull, not just push. But the sentiment data suggests the ceiling on goodwill is lower than revenue charts imply, and that any high-profile failure — a safety incident, a hallucinated medical answer, a deepfake election episode — will land on a public already primed to believe the worst. The labs are winning the usage war while losing the trust war, and the second one is the one that decides what the regulatory map looks like in 2027.
Working on something we should cover, or seeing a story we missed? Send leads, documents, or feedback to hello@aichatdaily.com. For sensitive tips, see our secure tips page for Signal and PGP options.
Spotted an error? Email hello@aichatdaily.com with the URL and the issue, or read our full corrections policy.
