Americans are embracing artificial intelligence, but with a suspicious eye
Artificial intelligence has entered the lives of Americans in a way few expected — fast, quietly, and at the same time, surrounded by doubt. A new survey from the Pew Research Center, published in June 2026, revealed a scenario that is curious to say the least: while the use of chatbots practically exploded over the past two years, skepticism about what this technology can do to society grew at the same rate.
It is that classic paradox — people use it, but they do not really trust what they are using.
And the standout finding is not just the growth in numbers. It is the fact that young people, who are exactly the ones using AI the most in their daily lives, are also the ones who most believe it will bring negative consequences for the future. So what is this survey really telling us? That the social impact of AI is no longer a theoretical discussion — it is happening right now, with millions of people, and the debate about its risks has barely started. 🤖
Chatbot growth by the numbers
The Pew Research Center survey showed that 49% of American adults now use chatbots like ChatGPT, Gemini, or Copilot. That number jumped dramatically — just two years ago, only a third of people had used this type of tool. About one quarter of Americans already turn to chatbots daily, something that would have been considered unlikely for such a recent technology. The growth was driven especially by ease of access, the fact that many of these platforms are free, and the natural curiosity that any new technology sparks — especially when it can hold conversations, create content, and solve problems in a surprisingly smooth way.
According to Anton Dahbura, an AI expert and co-director of the Johns Hopkins Institute for Assured Autonomy, that number is quite significant for the companies producing these products.
That kind of robust number suggests AI is officially mainstream, Dahbura said.
In terms of specific platforms, ChatGPT leads by a wide margin. More than four in ten Americans said they have used the OpenAI tool, making it the most widely used chatbot. Following behind are Google Gemini at 24%, Microsoft Copilot at 17%, and Meta AI at 14%. Tools like Elon Musk’s Grok, Anthropic’s Claude, and Character.ai each came in below 10% adoption.
But there is an important detail in this picture of accelerated adoption: using does not mean trusting. The same survey found that a significant share of regular chatbot users have serious concerns about the reliability of the information these tools generate. Factual errors, hallucinations — which is when AI simply makes up data with absurd confidence — and the lack of transparency about how these systems work are all factors that weigh heavily on how people feel. In other words, adoption grew, but the relationship between the user and the tool is still far from one of full trust. 📊
What Americans use chatbots for
The survey also broke down how these tools are being used in practice. Among respondents who use chatbots, the top use cases are:
- Searching for information: 42% of users turn to chatbots to research various topics
- Work tasks: 38% of those who are employed use AI for professional activities
- Entertainment: about 25% use chatbots for fun
- Creating or editing images and videos: a similar proportion to entertainment
- Medical advice or information about diet and fitness: about 20% of users
- News consumption: fewer than 15% use chatbots as a source of journalism
- Emotional support: 10% of Americans reported using chatbots for personal conversations
Dahbura raised a specific warning about that last category. He recommended caution regarding excessive dependence on and misplaced trust in chatbots, especially when it comes to emotional support. AI tools have no consciousness, no real empathy, and bear no responsibility for the well-being of those who use them.
Receive the best innovation content in your email.
All the news, tips, trends, and resources you're looking for, delivered to your inbox.
By subscribing to the newsletter, you agree to receive communications from Método Viral. We are committed to always protecting and respecting your privacy.
The profile of who uses and who does not
Usage patterns vary significantly depending on age group. The survey showed that two-thirds of Americans between 18 and 29 use chatbots, as do 61% of people between 30 and 49. Among those over 50, the majority still do not use these tools. Adults with a college degree tend to use chatbots more frequently and for more complex purposes, such as research, professional writing, and data analysis. Younger users engage with them in a broader and more varied way — from entertainment to emotional support to academic tasks.
This breakdown matters because it helps explain why skepticism and heavy use coexist so comfortably within the same generation.
Skepticism growing right alongside usage
One of the most intriguing findings from the Pew survey is precisely this seemingly contradictory coexistence of mass adoption and growing distrust. According to the data, 40% of Americans believe artificial intelligence will have a negative impact on society over the next two decades, compared to just 16% who see a positive impact. Another 31% believe the effect will be neutral, and the rest simply are not sure.
And among young people aged 18 to 29 — the ones who use chatbots the most on a daily basis — the pessimism runs even deeper. 48% of Americans in that age group expect AI to have a negative impact on society. That somewhat breaks the narrative that the more people use something, the more they trust it. In practice, it seems like the more people interact with AI, the more they notice its limitations and the risks it carries in the long run.
Older Americans are less likely to predict a negative outcome, but they are also more likely to feel uncertain about what lies ahead.
I think Americans are correctly recognizing that AI can be used for good or for nefarious purposes, Dahbura said. And it is smart for them to be concerned in the environment we are in right now, where it is basically anything goes.
The top concerns among Americans
The concerns raised by respondents revolve around well-defined themes. When it comes to employment, the fear that chatbots and artificial intelligence systems will replace human roles shows up consistently in the responses — and it is not an unfounded fear, considering that companies around the world are already restructuring teams precisely because of automation.
On the misinformation front, the concern is even more immediate. Dahbura pointed out that threats like deepfakes simply did not exist a few years ago and today already represent a concrete problem. With the ease of generating fake text, images, and even videos using AI, the potential for narrative manipulation at scale has never been higher.
And privacy rounds out the top three concerns. 71% of Americans expect AI to make their personal information less secure, according to the survey. Dahbura said Americans are being astute in recognizing these threats to data security.
Nearly two-thirds of respondents believe AI is advancing too fast. And that is also a legitimate concern in the absence of effective safeguards.
There are probably more side effects of AI to come that we have not even thought of yet, Dahbura added. 🔍
AI is happening to people, not just for them
Dahbura made an observation that sums up the current moment well. He pointed out that Americans are probably interacting with artificial intelligence far more than they realize, given the presence of this technology in household devices, phone calls, virtual assistants, smart thermostats, security cameras, and much more.
AI is happening to us, Dahbura said. That is, I think, the best way to put it.
The Pew survey confirms that perception by showing that 96% of Americans have heard or read something about artificial intelligence — an increase of 11 percentage points over the past four years. Awareness of what chatbots are specifically has also grown compared to two years ago. The topic has moved beyond niche territory and become part of everyday conversation.
Concerns about AI — whether at home, at work, or in schools — are no longer hypothetical, as Dahbura emphasized. The usage numbers from the survey show they are impacting millions of people in the here and now. Americans encounter AI without even trying, practically every single day.
The regulation question: trust is running low
If there is one thing the survey made especially clear, it is that Americans do not trust the government to regulate this technology effectively. More than two-thirds of respondents said they have no confidence in the government’s ability to regulate AI in a meaningful way.
And Dahbura considered that concern fair.
Yes, absolutely, he said. And especially when you take into account the influence of the AI industry, it does not look particularly good that consumers and individuals will be protected in the near future.
He noted that safeguards are insufficient, with only a handful of American states having taken the initiative to legislate on the matter on their own. This regulatory vacuum is concerning precisely because the technology keeps evolving, and the risks multiply at the same speed — or faster — than institutions can respond to them.
The social impact of AI is already a reality
The discussion about the social impact of artificial intelligence used to be treated as something for the future — a problem humanity would need to solve at some point down the road. But the Pew Research Center survey makes it clear that future has already arrived, and it showed up without much warning. With half of American adults having already interacted with some type of AI chatbot, the effects of this technology on behaviors, relationships, the job market, and even mental health are already observable in the present.
Healthcare professionals report patients who prefer to talk with chatbots before seeking actual medical care. Teachers deal daily with AI-generated assignments. Journalists compete with systems that produce content in seconds. This is the real, concrete impact already underway.
Another point that deserves attention is the issue of inequality in access to and benefits from AI. While users with higher education levels and income can extract real competitive advantages from these tools — whether to boost productivity or learn new skills faster — populations with less access to technology are left out of those gains. This creates a real risk of deepening inequalities that already exist. The conversation about democratizing AI, therefore, is not just a tech issue — it is deeply political and social.
It is also worth remembering that a recent report found that nearly 9 out of 10 children in the United States are already using some form of AI, raising concerns about educational and emotional dependence among younger generations. That is a development that adds yet another layer of complexity to the picture.
AI can be positive, but trust will take time
Despite the predominantly cautious tone, Dahbura acknowledged that artificial intelligence can be a net positive for society overall. He pointed to significant advances in areas like software development, medicine, and others that are already benefiting enormously from the technology.
But he does not expect Americans to start trusting AI anytime soon.
I think it is going to be a bumpy road ahead unless we adopt a strategic and thoughtful path, not just on a national scale, but really internationally, because the technology does not respect country borders, Dahbura said.
That statement captures the tension defining the current moment well. The technology is racing forward, the benefits are real, but so are the risks — and the infrastructure for governance, regulation, and public debate simply is not keeping up. 🧠
What this survey leaves us with
The picture the Pew survey paints is of a society that has already moved past the curiosity phase with artificial intelligence and is starting to grapple with the tough questions that come after. Usage grew, the tools improved, chatbots became more sophisticated — but skepticism also grew, and it is not born from ignorance. On the contrary, it seems like the more people get to know AI, the more they understand it needs boundaries, regulation, and a public conversation far more honest than what we have seen so far.
The social impact of this technology will continue to unfold over the coming years in ways we still cannot predict with precision. But one thing the survey already made very clear: the conversation about AI cannot stay confined to labs, tech companies, and experts. It needs to reach schools, communities, public policy, and most importantly, the hands of the people who are already using these tools every day — whether or not they fully realize what that means. 💡
