Cities Are Putting AI to Work, But Trust Still Lags

Cities are turning AI into civic infrastructure. A ThoughtLab benchmarking survey of 525 officials across 214 cities, supported by an analysis of 212 published city AI plans and interviews with more than 50 urban experts, shows a clear shift in ambition. The interviews covered 331 municipalities across every region and level of development.

City leaders no longer see AI only as a tool for cutting costs or speeding up paperwork. They see it as a way to make government more human by linking data across departments, replacing fragmented processes with coordinated action, and helping residents get support without navigating multiple disconnected systems.

That ambition has numbers behind it. Nearly two-thirds of city officials, or 64%, see AI as a key competitive driver, while 60% expect it to deliver more personalized, frictionless services. More than half aim to place people at the center of their AI strategy.

Nicole Coughlin, CIO of Cary, North Carolina, described the opportunity as a chance to “fundamentally rethink how government works.” The practical version is less cinematic: public employees spend less time on routine tasks and more time handling complex human problems.

Useful AI Looks Like Better City Services

Cities are applying AI across seven domains: government administration; safety and security; digital and physical infrastructure; energy and the environment; transport and mobility; health and education; and economy and industry. Adoption runs deepest where city operations are established, and thins out where success depends on complex coordination, specialized data, and equity-centered design.

The strongest examples are concrete. Seoul is piloting an AI chatbot that gathers emergency-call details by voice and flags the most urgent cases for operators. Honolulu’s CivCheck cut AI-assisted plan-review cycles by 58% and saves permit applicants more than 40 days on average.

Gold Coast uses predictive pipe maintenance to catch leaks before they become costly failures, while New York’s transit authority applies predictive maintenance to fix bus faults before they disrupt service. In Helsinki, Hester AI helps residents find online health and social-service guidance and aims to ensure support reaches the most vulnerable people in society.

The obstacles are less glamorous but more decisive. Eight in ten cities cite long-term costs, competing budget priorities, limited AI skills and talent, and difficulty justifying return on investment as major hurdles. A chatbot is easy to announce; proving that it improves public services without creating new gaps is the work.

Dubai Shows the Scale — And the Data Problem

Dubai is building its AI program around a broad digital strategy rather than isolated experiments. Launched in 2023, Dubai’s Digital Strategy aims to digitalise all aspects of life in the emirate and establish a system that strengthens the digital economy, while the D33 agenda is a 10-year plan focused on doubling Dubai’s economy by 2033.

The D33 agenda includes an estimated AED100 billion ($27.23 billion) in annual digital-transformation gains. In April this year, the government announced that its AI-driven initiatives should support more than AED10 billion ($2.72 billion) in GDP growth within two years of operation.

Digital Dubai introduced its AI Integration Matrix Framework in April this year to coordinate adoption across government entities and has applied it internally to guide the deployment of more than 100 AI systems. Its Dubai Population Now platform combines data from residency, transport, utilities, healthcare, property, education, tourism, and the public-sector workforce to estimate how many people are present in the emirate.

At the end of 2025, Dubai had 4.58 million residents, but its average daytime population peaked at 6.39 million. That additional 1.81 million people equals around 40% of the resident population — a useful reminder that cities serve the people moving through them, not just the people counted on a registry.

H.E. Younus Al Nasser, CEO of Dubai Data & Statistics Establishment at Digital Dubai, called data “the glue of our cities.” He said Dubai wants to combine information from different city verticals to create “one holistic picture” of how the emirate is performing.

That effort depends on data quality. In July, Dubai launched an updated Dubai Data Manual covering standards for data quality, governance, accountability, and compliance. Al Nasser put the problem plainly: “If we feed AI with unreliable data, then the answers we are getting out of it are not necessarily reliable, and that means we cannot take action with it.”

Dubai placed second globally in the Boston Consulting Group Intelligent Cities Index in August, which assessed 61 cities on their maturity in using digital technologies and AI to deliver tangible outcomes. BCG ranked Dubai first globally for adopting AI technologies and smart-city solutions.

Trust Is the Missing Deployment Layer

The corporate picture is less triumphant. Most companies Daria Rudnik works with now use AI every day, and McKinsey’s “State of AI in 2026” found that 80% said AI improved productivity. Only 37% attributed at least some EBIT impact to AI, while PwC’s “29th Global CEO Survey” found that 56% of CEOs had seen “neither revenue nor cost benefits” from AI.

The public trust gap is wider still. Only nine percent of Americans believe AI will do more good than harm, and over three hundred local communities have passed bans or moratoriums on hyperscale data centers over utility bills, noise, strained power grids, and limited local input. Texas Governor Greg Abbott has called for a pause in data center construction.

Regulation is not settling down to match the technology. In the US, the 2026 conversation has focused on deregulation: the White House has challenged state-level restrictions, federal agencies have withdrawn AI guidance, and policymakers have proposed reducing certain technology requirements. At the same time, the global library of AI-related laws, regulations, standards, and frameworks grew from around 250 instruments in March to 619 in September, with more than 200 coming from US states alone.

The pattern is clear. Cities can use AI to save time, coordinate services, and understand changing populations, but only when governance and trustworthy data arrive first. Otherwise, the grand future of intelligent government becomes another dashboard with excellent branding and questionable answers.

Stay connected via Google News

Follow us for the latest AI News