![\"A](\"https:\/\/alpha.pewresearch.org\/pewresearch-org\/wp-content\/uploads\/sites\/20\/2025\/03\/pi_2025.04.03_us-public-and-ai-experts_2-01.png?w=640\")
<\/a><\/figure>\n\n\n\nAI experts are far more enthusiastic than the American public about the increased use of AI in daily life. <\/strong>The public, on the other hand, expresses far more concern. Roughly half of the experts surveyed say they are more excited than concerned<\/em> (47%) about the increased use of AI in daily life. By contrast, only 11% of U.S. adults say this.<\/p>\n\n\n\n About half of U.S. adults (51%) say they are more<\/em> concerned than excited.<\/em> This drops dramatically to 15% among the experts surveyed.<\/p>\n\n\n\n What\u2019s more, the U.S. public has become more concerned over recent years. The share who say they are more concerned than excited increased from about four-in-ten in 2021 and 2022 to roughly half in 2023<\/a>.<\/p>\n\n\n\n Today, identical shares of both groups say that they are equally concerned and excited (38% each). <\/p>\n\n\n\n Among both the public and AI experts, men are more excited than women<\/strong> about the increased use of AI in daily life. The gender difference on excitement is wider among AI experts, though.<\/p>\n\n\n\n AI experts:<\/strong> A far greater share of men than women say they are more excited<\/em> than concerned (53% vs. 30%). While just 11% of men say they\u2019re more concerned<\/em> than excited, that ticks up to 24% of women.<\/p>\n\n\n\n U.S. public: <\/strong>Men are again more likely than women to say they are more excited than concerned (15% vs. 7%). While 46% of men are more concerned, that rises slightly to 55% among women.<\/p>\n\n\n\n In in-depth interviews, we asked AI experts about what uses of AI excite them, and why. Some themes include making life easier or more efficient and improving outcomes for certain industries. (Quotes have been lightly edited for grammar and clarity.)<\/p>\n\n\n\n \u201cI think broadly some of the things that excite me are things like applications that can save people a lot of time from repetitive and mundane tasks. So I think automating some of those workflows.\u201d<\/em><\/p>\n\n\n\n \u201cI\u2019ve seen that the AI can improve a lot the accuracy of the diagnosis of different diseases. Also, it can boost the development of different medicines for different treatments. Like for instance, for breast cancer classification, it can improve a lot. It can decrease the false positive rates and false negative rates. Most excited about the positive impact that it could have in the health industry.\u201d<\/em><\/p>\n<\/div>\n\n\n\n And we also asked experts what uses of AI concern them, and why. Some themes include data privacy and misinformation.<\/p>\n\n\n\n \u201cI do think about how that [airport biometrics] technology is used, especially from a privacy and security standpoint. … Where\u2019s that data going? How is it being housed? Where is it being used for? Where is my consent? Can I really, truly say no, I don\u2019t want my picture taken, but what is the consequence of me saying that and still trying to make it to my flight at home?\u201d<\/em><\/p>\n\n\n\n \u201cMisinformation has always been an issue with technology. \u2026 But I think the main issue with AI and misinformation is that you can now do misinformation at scale, at a way larger scale.\u201d<\/em><\/p>\n<\/div>\n\n\n\n A 2021 Center survey<\/a> found that some related themes also arose among U.S. adults when asked why they were either more concerned or more excited. <\/p>\n\n\n\n Our new survey also gives us the chance to compare expert and public concern in several key areas, including those related to \u201cdeepfakes,\u201d misinformation, job displacement and AI bias.<\/p>\n\n\n\n The public is more worried about losing jobs \u2013 and human connection \u2013 than AI experts. <\/strong>Continuing a theme from our broader body of research<\/a>, we find the public is anxious about AI\u2019s impact on work. More than half of U.S. adults are extremely or very concerned about AI eliminating jobs, versus a smaller share of experts surveyed (56% vs. 25%). The public also fears the loss of human connection more than experts do (57% vs. 37%).<\/p>\n\n\n\n There\u2019s wide concern about inaccurate information.<\/strong> Seven-in-ten of the experts we surveyed and 66% of U.S. adults are highly worried about people getting inaccurate information from AI.<\/p>\n\n\n\n Impersonation and data misuse are also among the top concerns. <\/strong>The public is more worried about each of these things than experts \u2013 most U.S. adults are highly worried. Still, six-in-ten experts say they are extremely or very concerned about data misuse, and roughly two-thirds say this about AI being used to impersonate people.<\/p>\n\n\n\n Experts and the public align in their concerns about bias.<\/strong> Identical shares of each group (55%) are highly worried about this. About half or more of experts and the public also express notable concern about people not understanding what AI can do.<\/p>\n\n\n\n There are gender differences on specific concerns about AI as well. Some of the biggest are on data misuse, bias and inaccurate information.<\/p>\n\n\n\n On the other hand, women feel similarly to men about impersonation and job loss.<\/p>\n\n\n\n Among the general public, most gender differences on this topic are minimal.<\/p>\n\n\n\n Loss of human connection is one place we see women being somewhat more concerned in both groups, though. Women are slightly more likely than men to be highly worried about AI leading to this, both among experts (45% vs. 35%) and the public (63% vs. 52%).<\/p>\n\n\n\n For some, concerns about AI extend to how they see their own futures.<\/p>\n\n\n\n The public is more likely to foresee personal harm from AI than benefit, though there is notable uncertainty:<\/p>\n\n\n\n In contrast, the AI experts we surveyed are widely optimistic:<\/p>\n\n\n\n Men stand out for thinking they\u2019ll benefit from AI in both groups we surveyed.<\/p>\n\n\n\n Among U.S. adults, men are more likely to foresee personal benefit than women (31% vs. 18%). This gender difference is also present among the experts surveyed (81% vs. 64%).<\/p>\n\n\n\n Many of the conversations around harm in AI center around potential biases related to race, ethnicity<\/a> and gender<\/a>. These include examples from hiring algorithms<\/a> to medical decision-making<\/a>.<\/p>\n\n\n\n One way AI companies are dealing with potential racial and gender biases in their models is by improving how models are trained<\/a>. At the same time, there are also calls for more diverse workforces<\/a> to counteract bias<\/a>. These efforts, though, have been met with pushback<\/a> as companies scale back these initiatives<\/a>.<\/p>\n\n\n\n We explored how the public and experts view the representation of six groups in this light, with questions we first asked in 2021<\/a>.[3. numoffset=”3″ Our survey focused on race, ethnicity and gender because these feature in prominent examples of AI bias<\/a> and due to constraints on survey space. However, there are a range of dimensions relevant to bias in AI.]<\/p>\n\n\n\n Both experts and the public believe White adults\u2019 views are better-represented than other groups\u2019 when it comes to AI design:<\/p>\n\n\n\n The public is more uncertain than experts with roughly four-in-ten or more unsure about these issues. But the views of White adults\u2019 representation again stand out:<\/p>\n\n\n\n We saw similar patterns in which groups stand out as well-represented the last time we asked the public about their views on this in 2021<\/a>.<\/p>\n\n\n\n Differences by race and ethnicity among the general public are mostly modest. Refer to Appendix B<\/a> for how these views break down.<\/p>\n\n\n\n Among experts, there were not enough Black or Hispanic experts surveyed to be broken out into a separate analysis \u2013 reflecting the racial and ethnic makeup<\/a> of the field as a whole. As always, these groups\u2019 responses are incorporated into the general figures for experts\u2019 views throughout this report.<\/p>\n\n\n\n Considerable attention has been paid to women\u2019s representation in the tech workforce \u2013 with many seeing them as underrepresented<\/a>.<\/p>\n\n\n\n Both the public and the experts we surveyed see the views of women as less well-represented than men\u2019s. Three-quarters of experts say the views of men are taken into account at least somewhat well by AI designers. In contrast, 44% say this about women\u2019s views.<\/p>\n\n\n\n Again, the public is notably uncertain \u2013 about four-in-ten adults are unsure how well these views are accounted for. Still, they see men\u2019s views as better accounted for than women\u2019s (42% vs. 27%).<\/p>\n\n\n\n By gender, among AI experts surveyed and U.S. adults<\/em><\/p>\n\n\n\n Women themselves are less optimistic than men when asked about the views of their own group.<\/p>\n\n\n\n Among experts, women are somewhat more <\/em>optimistic than men when asked about men\u2019s views (86% vs. 73% say these views are well-represented in AI). But there are no gender differences in this view among the public.<\/p>\n\n\n\n \u201cI identify as a woman, female, and I belong to a minority community. And I definitely think we need more representation, not just in terms of gender, but also in terms of ethnicities, places where people come from.\u201d<\/em> \u2013 Asian expert<\/em><\/p>\n\n\n\n \u201cI wanted [to use AI] to create an image of a Black family sitting around the table \u2026 Of course, there\u2019s many different blends of African American, and very different blends of Black people, but I thought it was interesting that it generated a series of images where everyone was biracial. If you don\u2019t know or if you can\u2019t relate to what a Black family looks like, it\u2019s only going to be amplified in the model that you’re creating, right? \u2026 Now, the interesting thing is it\u2019s only a reflection of the biases that are in the people who are creating the information \u2026 we all have unconscious biases.\u201d \u2013 Black expert<\/em><\/p>\n\n\n\n \u201cMost of the data that these large-scale AI systems are trained on is scraped from the internet. And [that] data \u2026 is inherently biased, not only according to those social categories that you mentioned, gender and race and so on, but \u2026 generally it comes from wealthy Western countries, wealthy White countries, and in particular along the coastline of those countries. \u2026 So it\u2019s a misconception that there is a malicious computer scientist embedding racism and sexism into the model. It\u2019s completely unintentional. It\u2019s essentially a product of ignorance of the data sources and the biases that are inherent in where we web-scrape the data from.\u201d \u2013 White expert<\/em><\/p>\n<\/div>\n\n\n\n We also asked experts and the public to consider if AI could perform and even improve upon jobs traditionally done by humans. In a number of cases, experts are more optimistic.<\/p>\n\n\n\n Experts are more likely than the public to say today\u2019s AI could outperform humans on many tasks \u2013 like driving, customer service and loan decisions. <\/strong>For example, 51% of the experts we surveyed say AI would do a better job driving people around; 19% of the public say the same. (They could also say AI would do worse, about the same or that they weren\u2019t sure.)<\/p>\n\n\n\n Few in the public think AI will outperform humans on any of the tasks we explored.<\/strong> The shares who say AI would do better than people whose job it is to do each task range from just 10% for parole decisions to 26% for medical diagnoses.<\/p>\n\n\n\n Both experts and the public doubt AI would best humans at certain tasks, including writing songs. <\/strong>Few among both the experts we surveyed (16%) and the general public (14%) think AI would do a better job in this case. Parole and hiring decisions<\/a> are also among the tasks ranking low on this list.<\/p>\n\n\n\n Still, some of these views are mixed overall, even among the experts we surveyed. And the public is often more unsure than experts. For example, when asked about customer service, 16% of U.S. adults and 4% of experts say they\u2019re not sure who would do a better job.<\/p>\n\n\n\n Lawmakers are debating who should regulate AI<\/a> \u2013 and to what extent. We explored several aspects of these debates in our surveys.<\/p>\n\n\n\n Experts and the public are more worried government regulation will be too lax:<\/strong> The public and experts surveyed both more commonly say they are concerned that the U.S. government will not go far enough in regulating AI, rather than going too far.<\/p>\n\n\n\n At the same time, 21% of experts and 16% of U.S. adults say they\u2019re uncertain about this. (The surveys were fielded in summer and fall 2024, before the change in presidential administration.)<\/p>\n\n\n\n Majorities of U.S. adults in both parties are more concerned about insufficient regulation.<\/p>\n\n\n\n Still, Democrats (64%) are more likely than Republicans (55%) to say this. (Both groups include those who lean toward the respective party throughout this report.)<\/p>\n\n\n\n Experts and U.S. adults are largely worried about insufficient regulation. But they\u2019re also not convinced that the U.S. government can make it happen.<\/p>\n\n\n\nBy gender, among AI experts surveyed and U.S. adults<\/h5>\n\n\n\n
Quotes from AI experts: Reasons for excitement about AI<\/h4>\n\n\n\n
Quotes from AI experts: Reasons for concern about AI<\/h4>\n\n\n\n
Specific concerns about AI<\/h4>\n\n\n\n
![\"A](\"https:\/\/alpha.pewresearch.org\/pewresearch-org\/wp-content\/uploads\/sites\/20\/2025\/03\/pi_2025.04.03_us-public-and-ai-experts_2-02.png?w=640\")
<\/a><\/figure>\n\n\n\nBy gender, among AI experts surveyed and U.S. adults<\/h5>\n\n\n\n
![\"A](\"https:\/\/alpha.pewresearch.org\/pewresearch-org\/wp-content\/uploads\/sites\/20\/2025\/03\/pi_2025.04.03_us-public-and-ai-experts_2-03.png?w=640\")
<\/a><\/figure>\n\n\n\nAI\u2019s personal impact<\/h3>\n\n\n\n
![\"A](\"https:\/\/alpha.pewresearch.org\/pewresearch-org\/wp-content\/uploads\/sites\/20\/2025\/03\/pi_2025.04.03_us-public-and-ai-experts_2-04.png?w=620\")
<\/a><\/figure>\n\n\n\n\n
\n
By gender, among U.S. adults and AI experts surveyed<\/h4>\n\n\n\n
<\/a><\/a>Representation in AI design<\/h3>\n\n\n\n
How well are the views of racial and ethnic groups represented?<\/h4>\n\n\n\n
![\"A](\"https:\/\/alpha.pewresearch.org\/pewresearch-org\/wp-content\/uploads\/sites\/20\/2025\/03\/pi_2025.04.03_us-public-and-ai-experts_2-05.png?w=620\")
<\/a><\/figure>\n\n\n\n\n
\n
By race and ethnicity, among U.S. adults and AI experts surveyed<\/h5>\n\n\n\n
How well are the views of men and women represented?<\/h4>\n\n\n\n
![\"A](\"https:\/\/alpha.pewresearch.org\/pewresearch-org\/wp-content\/uploads\/sites\/20\/2025\/03\/pi_2025.04.03_us-public-and-ai-experts_2-06.png?w=640\")
<\/a><\/figure>\n\n\n\n\n
Quotes from AI experts: Representation and bias<\/h4>\n\n\n\n
AI versus humans<\/h3>\n\n\n\n
![\"A](\"https:\/\/alpha.pewresearch.org\/pewresearch-org\/wp-content\/uploads\/sites\/20\/2025\/03\/pi_2025.04.03_us-public-and-ai-experts_2-07.png?w=640\")
<\/a><\/figure>\n\n\n\nRegulation and responsible use of AI<\/h3>\n\n\n\n
Government regulation<\/h4>\n\n\n\n
![\"A](\"https:\/\/alpha.pewresearch.org\/pewresearch-org\/wp-content\/uploads\/sites\/20\/2025\/03\/pi_2025.04.03_us-public-and-ai-experts_2-08.png?w=640\")
<\/a><\/figure>\n\n\n\nBy political party, among U.S. adults<\/h5>\n\n\n\n
Confidence in government and industry<\/h4>\n\n\n\n
![\"A](\"https:\/\/alpha.pewresearch.org\/pewresearch-org\/wp-content\/uploads\/sites\/20\/2025\/03\/pi_2025.04.03_us-public-and-ai-experts_2-09.png?w=640\")
<\/a><\/figure>\n\n\n\n
![\"A](\"https:\/\/alpha.pewresearch.org\/pewresearch-org\/wp-content\/uploads\/sites\/20\/2025\/03\/pi_2025.04.03_us-public-and-ai-experts_2-10.png?w=640\")
<\/a><\/figure>\n\n\n\n