New research from CTG suggests public libraries can play an important role in helping communities develop the skills to critically evaluate the technology. (Photo by Brian Burke/CTG)
ALBANY, N.Y. (Sept. 24, 2026) — As artificial intelligence tools become increasingly embedded in work, government and everyday life, new research from UAlbany’s Center for Technology in Government (CTG) suggests public libraries can play an important role in helping communities develop the skills to critically evaluate the technology, understand its risks and participate in decisions about how it is used.
The study, “Who Participates in AI Programs at Public Libraries? Exploring Demographic and Attitudinal Factors,” was presented at the 18th International Federation for Information Processing Working Group 8.5 International Conference on Electronic Participation, or ePart 2026, held in Athens, Greece. The paper, authored by Zong-Xian (Sam) Huang, J. Ramon Gil-Garcia and Mila Gascó-Hernandez, was published in the conference proceedings.
The research examines not only who participates in AI-related programs at public libraries, but how participants’ perceptions of the benefits and risks of AI relate to their likelihood of taking part in such programs.
Who is participating?
Drawing on a January survey of 2,010 U.S. adults, the CTG researchers found significant differences in participation based on age, gender, education and income.
Compared with respondents ages 18 to 34, those 55 and older were associated with a 32.1-percentage-point lower probability of participating in library AI programs. Women were 9.8 percentage points less likely than men to participate.
Education and income were also associated with participation. Respondents with at least a four-year college degree were 7.1 percentage points more likely to participate than those without one, while respondents from households earning more than $100,000 annually were 9.4 percentage points more likely to participate than those earning less than $50,000.
Another notable finding concerned race. Non-white respondents were associated with a 10.4-percentage-point higher probability of participation than white respondents, a result the researchers said warrants further study.
“What stood out to us is that participation is not evenly distributed. Some of the groups that may face greater barriers to engaging with new technologies are also less likely to take part in the AI-related programs offered by public libraries,” said Huang, a PhD candidate in UAlbany’s Rockefeller College of Public Affairs and Policy and graduate research assistant at CTG. “Public libraries can provide an inclusive, accessible setting where people can expand their knowledge, ask questions and become more confident navigating a rapidly changing digital world.”
The implications extend beyond who attends a particular library program. If participation in AI education is uneven, the researchers argue, opportunities to develop the knowledge needed to engage in broader civic discussions and decisions about AI could become uneven as well.
The value of constructive skepticism
The authors also found that having concerns about AI did not necessarily discourage participation. Respondents who expressed specific concerns, including concerns about whether the public has enough expertise to weigh in on AI, whether public feedback would actually be used and some potential privacy issues, were more likely to participate in library AI programs than those who expressed broader concerns across multiple areas.
The researchers describe these participants as “constructively skeptical,” suggesting that specific questions or concerns about AI may give people a reason to learn more and become involved.
“AI literacy is about more than knowing how to use these tools,” said Gascó-Hernandez, research director at CTG and a professor in Rockefeller College’s Department of Public Administration and Policy. “People also need to understand their limitations, think critically about the information they produce and recognize issues involving accuracy, privacy and bias. Libraries can help people build those skills so they can make informed choices about when and how to use AI.”
The research builds on CTG’s examination of how public libraries can help communities understand artificial intelligence and participate in decisions surrounding its use. Earlier research examined how libraries are developing AI-related programs and their potential role in addressing emerging digital divides. The new study shifts the focus toward identifying who is actually participating and which populations may be underserved.
“These findings give us a better picture of who is participating, but they also raise important questions about how libraries can reach different communities more effectively,” said Gil-Garcia, director of CTG and a professor in Rockefeller College. “The next step is to better understand what kinds of programs, partnerships and outreach can bring more people into these conversations and help libraries design AI initiatives that respond to the needs of the communities they serve.”
The researchers suggest that public libraries consider partnerships with community organizations and more targeted programming to reach populations that are less likely to participate. They also recommend connecting the broader benefits of AI engagement to tangible ways the technology affects individuals as well as creating opportunities for patrons to raise critical questions about its use.
The project was supported by the Institute of Museum and Library Services.
