As artificial intelligence (AI) tools become increasingly prevalent within public sector frameworks, a new study from the Center for Technology in Government at the University at Albany (CTG UAlbany) investigates the practical application of these technologies. The research specifically examines how government agencies are deploying chatbots and evaluates the tangible changes these tools are introducing to administrative practices. By moving beyond theoretical discussions, the study provides a grounded perspective on the current state of AI in governance.
Empirical analysis of artificial intelligence integration in public administration
The study, titled “Uncovering the Results of AI Chatbot Use in the Public Sector: Evidence from US State Governments,” was co-authored by Tzuhao Chen and Mila Gasco-Hernandez. The findings are based on extensive interviews conducted with officials from 22 state agencies across the United States. This empirical approach offers a rare and detailed look at how chatbot technology is actively influencing both internal governmental operations and the quality of interactions with the general public.
Tzuhao Chen, an associate professor of public policy and administration at Florida International University, emphasized the necessity of this shift in focus. He observed that while much of the existing discourse regarding artificial intelligence in public administration has centered on future potential, there is a critical need to understand the actual experiences of organizations that have already integrated these tools. Chen, who earned his doctorate from the Rockefeller College of Public Affairs and Policy in 2025, previously served as a graduate researcher at CTG UAlbany, bringing a deep academic and practical background to the investigation.
The research indicates that the implementation of chatbots is doing more than just automating simple tasks; it is fundamentally reshaping how agencies manage information and respond to constituent needs. By analyzing the evidence from state governments, the authors illustrate that these tools are becoming essential components of modern administrative infrastructure. The study highlights that the successful use of AI depends largely on how well these tools align with the specific operational goals and constraints of each individual agency.
Furthermore, the evidence suggests that the use of chatbots has a direct impact on the efficiency of public service delivery. By handling routine inquiries, these AI systems allow government personnel to dedicate more time to complex issues that require human intervention. This shift not only optimizes internal workflows but also changes the expectations and experiences of citizens when they engage with state authorities. The study serves as an essential resource for policymakers looking to navigate the complexities of digital transformation in the public sphere.
Functional applications and operational impacts of chatbot integration
Within various public agencies, chatbots are primarily deployed to address routine inquiries and distribute essential information regarding governmental programs and services. This strategic allocation of automated resources allows administrative personnel to pivot their focus toward complex cases that necessitate human judgment, nuanced problem-solving, or direct personal interaction. By filtering high-volume, repetitive tasks through AI interfaces, organizations aim to optimize the utility of their professional staff.
Despite the clear intent behind these deployments, the study indicates that the definitive impact of chatbots on overall institutional workloads remains ambiguous. While several agencies reported a perceptible decrease in traditional communication volumes, such as telephone calls and emails, others observed a mere migration of demand across different digital channels. This shifting pattern of engagement makes it challenging for administrators to isolate and quantify the measurable advantages of automation in terms of net labor reduction.
Furthermore, the research highlights a more consistent trend regarding the qualitative shift in how agencies perceive and engage with their constituents. Rather than simply reducing the number of interactions, chatbots have altered the nature of the data collected from the public. This evolution in communication suggests that the value of AI in the public sector may lie as much in data acquisition as it does in immediate task completion.
The analysis of chatbot interactions provides organizations with sophisticated insights into user behavior, specifically regarding the terminology citizens employ and the common points of confusion they encounter. In numerous instances, the data harvested from these AI dialogues has compelled agencies to revise their website content and restructure how information is presented to the public. By identifying gaps or ambiguities in their existing digital resources, departments can proactively refine their communication strategies to better serve the community.
Mila Gasco-Hernandez, the research director at CTG UAlbany and an associate professor at Rockefeller College, characterizes chatbots as essential instruments for organizational learning. She posits that these tools do more than provide a service; they act as a diagnostic lens through which agencies can better understand the specific needs and linguistic patterns of the populations they serve. This iterative process of learning and adaptation ensures that public administrations remain responsive to the evolving requirements of a digital-first society.
Enhanced accessibility and operational responsiveness in public services
The integration of chatbot technology within public agencies has led to demonstrable improvements in information accessibility and institutional response times. By providing immediate automated assistance, these tools significantly reduce the necessity for traditional email exchanges or telephone consultations, which often involve lengthy wait times. Furthermore, chatbots function as a vital supplementary communication channel, catering specifically to constituents who prefer autonomous information retrieval or require assistance outside of standard administrative operating hours.
For modern government agencies, the implementation of artificial intelligence represents a strategic expansion of service delivery rather than a wholesale replacement of established methodologies. Chatbots operate in parallel with traditional human-led channels, effectively managing fluctuating demand while maintaining diverse points of access for a varied public demographic. This hybrid approach ensures that while automation handles high-frequency inquiries, the structural integrity and availability of legacy support systems remain intact for those who require them.
The data generated through these digital interactions has emerged as one of the most consequential assets for public administrations. Many organizations now utilize integrated analytical tools to monitor engagement patterns, identify frequently unanswered questions, and track user behavioral trends. This continuous feedback loop supports the ongoing optimization of both the chatbot interfaces and the underlying information architectures. In several instances, these data sets have highlighted significant discrepancies between formal bureaucratic terminology and the natural language used by the public, allowing agencies to refine their communications for greater clarity and usability.
As many chatbot initiatives remain in their foundational stages, the research indicates that evidence regarding their long-term systemic impact is currently limited. Observations to date primarily focus on short- to medium-term benefits, such as streamlined internal processes, accelerated query resolution, and broader public access to government resources. These incremental improvements suggest a trajectory toward a more efficient and user-centric model of public administration.
As agencies continue to scale the application of chatbot technology across a wider array of programs and services, researchers anticipate that these tools will occupy an increasingly central role in governance. The evolution of this technology is expected to fundamentally redefine the interface between public administrations and the citizens they serve, fostering a more responsive and technologically adept civil service. The ongoing commitment to refining these AI-driven interactions underscores a broader movement toward digital transformation within the public sector.
The study is published in Nature Ecology & Evolution.
r
