49% of Americans already use AI frequently. Only 16% believe that it brings something good

People are using AI non-stop, but they are afraid of it. It is the great conclusion of a recent study from Pew Research Center in which a unique dilemma is posed. Adoption is skyrocketing, but so is pessimism and distrust regarding AI.

A survey with contradictory results. The well-known Pew Research Center published its macro survey “Americans and AI 2026” these days, and the results give a very contradictory picture. On the one hand, generative AI tools are increasingly used and are already part of many people’s daily lives. On the other hand, this success in popularity coexists with the concern that these same users have regarding this technology.

ChatGPT is king. The report shows how using generative AI models is already very normal among the North American public. 49% of American adults say they use chatbots like ChatGPT, Gemini or Copilot, up from 33% in 2024. ChatGPT’s market share is by far the most notable: of all those who use these platforms, 44% (almost half) use ChatGPT over other alternatives. Gemini is in second position with a 25% share, and behind it are Copilot and Meta AI. The use is also anecdotal: 12% confess to using AI several times a day, and for 4% the use is “almost constant.”

Goodbye search engine, hello AI. The practical use of AI also allows us to understand the change that is taking place. In the survey, participants indicated that the two areas in which AI is most used are the search for information (40%) and the automation of work tasks (38% of active workers). Not only that: the AI ​​Overviews of the Google search engine are an absolute success: 60% of users take advantage of these summaries, which reveals a total transformation of the way we consume content and, in this case, searches.

It is striking that AI is being used more and more, but at the same time users have an increasingly negative perception as well.

Fear of AI (especially among young people). One of the most surprising conclusions of the study debunks the idea that new generations embrace AI with much more optimism than previous generations. It is true that those under 30 years of age double the AI ​​use rate of those over 50 (57% compared to 28%). However, it is precisely these young people who show a more hostile profile towards the impact of this technology. These young people, those most impacted by recommendation algorithms, are the same ones who overwhelmingly believe that AI will end up having a negative impact on both their personal lives and society in the next 20 years.

Less data security. The main reason for this negative perception of AI seems to be data security. For 71% of respondents, the use of AI will make their data less secure. It is a striking discovery that reveals a real problem: the conversations we have with AI are not private, because when asking something to ChatGPT, Claude or Gemini, for example, the data is sent to the clouds of OpenAI, Anthropic or Google (respectively) to be read and processed.

This data can theoretically be collected, stored and consulted by personnel from these companies, and encryption practically does not exist. Only solutions like Apple’s Private Cloud Compute They propose a real solution to the privacy and confidentiality of conversations with AI models. Google, by the way, has launched its own alternative, Private AI Compute.

A problem in the making according to US users. The growing use of AI seems positive because it shows that users are indeed finding ways to take advantage of it, but the fact that this increased use does not contribute to a positive perception is especially worrying. Here it is evident that the current rejection of data centers in the USA or fear of the impact that AI will have on work They do not help to improve that perception, and the uncertainty in both senses is high.

There is more concern about AI in Western countries: curiously in Asia, optimism exceeds the concern generated by this technology.

And in Spain, what? In our country the survey “AI Monitor 2026” carried out by Ipsos reveals varied data for various countries, including Spain. According to this survey, 47% of the Spanish population is enthusiastic about the use of AI (8 points above the European average), although there is also concern regarding this technology: 52% of those surveyed showed concern about its impact on society.

Image | Solen Feyissa

In Xataka | Spain has just put numbers to the impact of AI on the labor market: 2.3 million jobs will change forever