What Cities Keep Getting Wrong about AI implementation

Smart cities have long promised a future where technology seamlessly optimizes urban life, from reducing traffic congestion to improving air quality. But according to a new comprehensive review published in Smart Cities, the gap between the potential of artificial intelligence (AI) and its practical implementation remains significant, driven largely by overlooked ethical concerns, infrastructure gaps, and mismatched priorities.

The study, conducted by Radosław Wolniak and Kinga Stecuła at the Silesian University of Technology, analyzed 157 recent academic papers detailing AI applications across six key smart city sectors: governance, economy, mobility, environment, living, and people. It identified both groundbreaking advances and substantial barriers, highlighting essential lessons for cities aiming to leverage AI.

AI Successes and Shortcomings

AI is indeed transforming urban landscapes. In the mobility sector, AI-driven traffic management systems and autonomous vehicles are effectively reducing congestion and emissions. Smart governance, leveraging chatbots and automated decision-making, is enhancing citizen engagement and administrative efficiency. Smart living innovations such as AI-enabled healthcare monitoring and intelligent home management systems promise significant quality-of-life improvements.

Yet despite these successes, the study reveals cities are often failing to address critical barriers—ethical challenges, privacy concerns, and unequal access—that significantly hinder AI adoption.

AI is more than a technological upgrade

The research underscores a recurring oversight: Many cities treat AI primarily as a technological upgrade rather than a deeply social intervention. For instance, smart governance applications often neglect transparency, inadvertently fostering citizen distrust. Similarly, in smart economy initiatives, AI-driven automation sparks fears of job losses without clear plans for workforce transition.

Moreover, while smart mobility solutions like autonomous vehicles receive significant investment, ethical and privacy implications—such as the extensive personal data collected by these systems—often remain unaddressed. As a result, public backlash and regulatory challenges slow down otherwise promising technologies.

AI Amplifies Inequities, Unless Carefully Managed

Critically, the study highlights that AI implementations frequently exacerbate the digital divide. Wealthier, digitally literate communities disproportionately benefit, while marginalized populations face new forms of exclusion. This imbalance risks transforming smart cities into enclaves of privilege rather than engines of inclusive progress.

To prevent this, the researchers emphasize the necessity of proactive, inclusive policymaking that ensures equitable access to technology and digital education across all demographics.

Learning from the Research: A Roadmap for Smart Cities

So, what should smart cities take from this extensive review?

First, policymakers must embed transparency and ethical AI practices into every initiative. Transparent communication about AI decisions can mitigate distrust and increase public acceptance. Ethical guidelines should be foundational, not afterthoughts.

Second, cities must commit to extensive digital literacy programs. Ensuring all residents, regardless of age or socioeconomic status, have access to and understanding of AI technologies is essential to close the digital divide.

Third, AI should always be human-centered. Technologies need to adapt to human needs and community values, not vice versa. Cities must avoid the temptation of deploying flashy technologies without meaningful benefits or clear strategies for social inclusion.

Finally, cities should utilize AI for proactive, anticipatory governance. Predictive analytics can help authorities address challenges before they escalate—be it traffic management, environmental crises, or public health threats. Yet, this requires both investment in robust data infrastructure and responsible data governance practices.

The Future: Inclusive, Ethical, and AI-Enhanced

AI offers immense potential to revolutionize urban life positively. Yet, the critical insight from Wolniak and Stecuła’s research is that smart cities will only succeed if they deliberately address the social, ethical, and equitable dimensions of AI integration.

Cities that learn these lessons will not merely be "smart" in a technological sense; they'll also be wise, inclusive, and prepared for the complex societal transformations AI will inevitably bring.