Balancing Tech Advancements and Resident Trust: AI in the Public Sector
Real-world use cases, consumer sentiments, and what’s next for AI in government technology
We’re watching AI progress in real time
At this point, there’s little doubt that artificial intelligence (AI) is reshaping the entire digital landscape, including government technology. While there was some skepticism (and possibly surprise) early on in the AI boom, as public sector pros have had more exposure, their concerns have started to subside. Government agencies, traditionally known for slower adoption of new technologies, have been testing use cases, deploying AI tools quickly, and creating policies to mitigate risks.
While AI has been reshaping the way agencies operate, most IT leaders have emphasized the human aspect of their work. Incorporating AI isn’t about replacing employees. Leaders are optimistic about AI’s ability to automate processes, increase efficiency, and optimize data analysis for public benefit, but the goal is to improve government services and response times without losing the human touch.
How are government agencies using AI?
Within the past few years, public sector leaders’ assessment of artificial intelligence has drastically shifted — from AI being something interesting to keep an eye on to real enthusiasm about the possible efficiency benefits.
Back in 2023, AI didn’t even make it into the NASCIO State CIO Top Ten Priority Strategies and Solutions, and as a tool, it only placed 6th. But in 2025, NASCIO members vaulted AI to the number 2 spot for strategies and number 1 priority tool or application, reflecting the technology’s potential to address numerous challenges.
In 2025, we asked government leaders about the status of AI in their agency:
Are government agencies using AI already?
Among agencies that do not already use AI:
- 13% have implementations underway for back-office processes
- 15% are implementing a solution for a front-office process
- Another quarter of agencies are actively evaluating AI solutions — and only 2% reported they have no plans to deploy AI
Our study indicates that local (city and county) jurisdictions seem slightly ahead of the curve: 64% of local respondents said they’re already deploying AI, compared to 53% at the state level. Naturally, state-level AI deployments can be more complex than those for a city or county.
Our results mirror similar recent studies: Roughly half of state and local governments across the U.S. are already using AI. A 2024 Digital Cities Survey from the Center for Digital Government found that almost half (44%) of responding cities have AI-powered chatbots today, and another 44% say they’re coming soon.
A 2025 poll of state IT leaders by the National Association of State IT Directors (NASTD) found that 20% of the responding states have already incorporated AI into resident-facing services, and 41% are actively working on resident-facing AI projects. In the states that have already implemented consumer-facing AI tools, over 95% said they’re using AI-powered chatbots. Help center solutions and fraud detection were also popular use cases.
We can’t just take gigantic risks with AI, but we also are not going to put our heads in the sand. If you decide not to learn anything by the time AI becomes required, we won’t know anything about it because we won’t have done any work with it, and we won’t have any experience with it. So it was a mandate for us to begin the AI journey with publicly available data.
As training increases, people are more comfortable using AI
Leaders in both the private and public sectors are enthusiastic about adopting AI. In the private sector, C-suite respondents were the biggest daily users of AI (around 72% versus 45–56% for other roles). They also reported the highest impact and productivity increases compared to other roles, using AI as a path toward more opportunity. And although there’s persistent concern that workers will be replaced by AI, recent surveys suggest that public sector employees are getting more comfortable with AI.
With a focus on upskilling, public sector IT leaders are driving big changes year-over-year
In early 2024, only 13% of public sector organizations said they were using AI on a daily basis, and 60% of public sector IT professionals cited a lack of knowledge and skill as their top challenge to implementing AI.
However, by late 2024, over 70% of public sector staff said they’d received AI training within their agency, and 71% of those who have participated in such training said it was effective. By 2025, 78% of states reported that they’d created an AI advisory committee or task force and cited AI workforce expertise and training as a top consideration for developing an AI roadmap.
With the increased training and more firm AI policies taking shape, government leaders reported overwhelming optimism about AI in late 2024, with 89% of respondents expressing that AI has the potential to improve operations in state and local government over the next five years. Their enthusiasm bubbled over into 2025: 94% of state and local IT and program managers said they foresee increased AI usage in their agencies over the next one to two years. And in the NASCIO State of the States Tech Forecast 2025, over 70% of respondents said generative AI and machine learning will be the most impactful emerging IT technology over the next 3-5 years.
How early adopters are successfully using AI
Wake County, North Carolina, like many jurisdictions, started their journey into AI with low-risk experimentation using public data. This strategic but cautious approach protects sensitive resident data and allows government agencies to start exploring AI and develop governance policies.
Here’s how government agencies are using AI to improve resident services and optimize internal processes:
Chatbots
Provide residents and staff with fast and responsive information, 24/7, in dozens of languages.
- Utah County implemented chatbots in its emergency management system to collect and respond to resident input
- Researchers have developed GRASP, a purpose-built chatbot to assist residents in understanding municipal budgets
Pattern recognition
AI can (very quickly) parse data or text and point to patterns, so agencies can more easily identify traffic bottlenecks, monitor public health, and predict infrastructure malfunctions.
- In Utah County, UT, AI tools help government employees search internal HR policies more easily get faster, clearer answers
- Atlanta is using AI to detect leaks in water infrastructure
- Philadelphia is implementing AI-enabled cameras on school buses and public transit to detect traffic violations
Speech recognition, language processing, and text generation
AI can interpret audio, handwritten text, or a prompt and generate a digital version in minutes.
- Wake County, North Carolina, uses AI to transcribe handwritten records for projects like the Enslaved Persons Project
- Utah County Commissioner Amelia Powers-Gardner uses AI to generate first drafts of her speeches to improve efficiency
And while many leaders are enthusiastic about AI (or feel under pressure to implement this new tool), ignoring customer concerns about AI use could mean losing resident trust.
Residents are skeptical about AI in government
As with most emerging technologies, customers are still skeptical about the use of AI, especially in industries that handle sensitive personal information like health care and government services. Overall, more Americans have negative than positive views of AI’s potential impact, according to a 2025 Gallup poll; people cited being concerned about false information spreading, lost job opportunities, security risks, and diminished social connections.
In our 2024 consumer survey, 56% of respondents said they’re at least somewhat comfortable with government agencies using AI, but 44% of respondents said they are somewhat to very uncomfortable with this new technology. Avid digital users are more likely to be open to AI (66% of more digitally savvy people expressed comfort with AI in our study), but even the more tech-forward consumers report some hesitation.
Overall, trust in AI remains shaky at best. An April 2024 YouGov poll found that 62% of people don’t trust AI to make ethical decisions, and 45% don’t trust that AI’s information is accurate. Although respondents recognized that while AI could have benefits, the lack of regulation and potential for error gives them pause (along with other worries like scams, bugs, misinformation, and the human cost of AI, including job losses)
Quick takeaways
- Most U.S. adults, regardless of political party, believe the government should play a role in addressing the potential harms and risks associated with AI
- However, the majority of U.S. adults don’t think the government will do enough to regulate AI
- Consumers (who have been inundated with new “AI” features or products) have cited chatbot fatigue
- The majority of Americans think AI will become another obstacle between them and a customer success agent (i.e., AI will make it harder to talk to a real person)
- Similarly, 57% of the public is highly concerned about AI leading to less connection between people
- Professionals who work on AI are more enthusiastic about AI than the public
Weighing the risks
Despite enthusiasm about AI, government leaders retain some healthy skepticism. In 2024, research by the National Association of State IT Directors respondents cited concerns like higher security risk, underdeveloped solutions, and workforce worries.
Just because there’s an AI tool available doesn’t automatically mean government agencies should use it. AI introduces new risks: from failure to deliver required services to reputational risks if the implementation fails to meet expectations. Questions around bias, the fairness of models, and hallucinations are also valid concerns. The majority of state IT leaders have wrapped responsible AI use (like appropriate guardrails, data privacy, and cybersecurity) into their AI 2025 roadmaps.
We’re seeing a lot of impact with a copilot approach. So it’s not ‘let’s let AI run free,’ but how does it augment what’s being done? There’s still human oversight.
Best practices for putting AI into action
A good starting point for identifying AI use cases is to examine how AI is currently being implemented successfully in the public and private sectors. Like the examples we highlighted earlier, you can start with lower-risk AI implementations to improve day-to-day communication with the community:
- 24/7 customer support, via chatbot: AI-powered chatbots can answer common resident questions at any time without overworking staff or increasing headcount. Since residents will have more access to consistent answers about services like utilities, permits, licensing, or court dates, AI chatbots could help reduce the number of support requests your agency receives.
- Personalization: You can use AI to analyze resident behavior and engagement data to deliver more relevant messages like important deadline reminders, targeted alerts during emergencies, or service outage updates based on location.
- Translations: Real-time translation makes government services more accessible for people with limited English proficiency.
- Analyze resident feedback and comments quickly, at scale: AI tools can quickly review thousands of comments, emails, or survey responses to identify key themes, concerns, and sentiments. This helps leaders prioritize improvements based on real-time input.
- Fine-tune written communications: With the right prompts, AI tools can draft or review your agency’s written materials. Make sure the prompt includes clear instructions; ask it to use plain, concise language so your messages are consumable for a wide audience. You can also ask it to be more professional or friendly, depending on the type and urgency of the communication.
Make sure your agency is prioritizing regular training, ongoing support, and clear agency policies regarding AI. Institute general best practices:
- Build continuous monitoring and reevaluation into your AI roadmap
- Create thorough, specific, and clear AI policies
- Understand the security and privacy implications, and implement the proper protections
- Communicate transparently with residents and staff about the solution
- Start with narrow use cases and expand from there, building on success
We’ve created an AI user group; both state and local government employees can join. In just a little over a year that we’ve done this, more than 700 people have signed up, which speaks to the need. Folks share best practices, templates, policies, and those sorts of things.
How to talk about AI: Communicating changes to staff and residents
Approach AI communication thoughtfully, both internally with staff and externally with residents. Messaging should be proactive and transparent, covering goals, safety measures, and timelines.
Internal communication
for agency leaders integrating AI into broader digital transformation strategies:
- Apply a structured change management framework (why AI is being adopted, the impact on workflow, how job duties will change)
- Develop agency-wide AI use policies and share updates regularly
- Invest in staff training: online courses, in-person workshops, and cohorts led by peers or partners
- Encourage the IT team to give regular updates on how AI is integrated into the agency’s tech stack
- Create an AI task force with diverse staff representation, and encourage a feedback loop
External communication
for resident education campaigns:
- Provide regular updates via dedicated webpages explaining how AI is being used, videos or infographics that explain machine learning or predictive analytics, and press releases about AI use in local government technology
- Highlight resident impact like faster services, 24/7 support, or improved infrastructure monitoring
- Openly address risk and ethical concerns, and publish your AI use policies
- Start with small pilot projects before making any large changes with AI
- Consider inviting public comment on draft policies or pilot projects or partnering with local universities or civic tech groups to validate models and reduce bias
With a clear communication strategy around AI use, leaders can include both staff and residents in the modernization journey — building support, reducing friction, and setting the stage for responsible innovation.
The right balance between trust and innovation
Government and technology both exist to serve people. AI can help, if it’s used carefully and responsibly. And we’re only at the beginning of what AI will be able to achieve. But as government agencies increasingly explore AI to modernize services, they face a dual imperative: harness AI’s potential while maintaining public trust and mitigating risk. With sharp consumer skepticism, it’s critical for government leaders to focus on reducing risks before deploying AI. Forward-looking agencies are adopting a cautious but strategic approach — with transparent frameworks, public feedback, and human oversight for high-impact decisions. Ultimately, resident trust isn’t earned through technology alone, but through thoughtful AI implementation, ethical policies, and clear communication.
