AI could make smart cities more liveable and responsive to community needs

A genuinely smart city must keep offering accessible human services and ensure technological progress does not deepen existing inequality, writes Paul Budde.

FOR MANY YEARS, I have argued that a smart city is not defined by the amount of technology it deploys. A city becomes smarter when it uses technology to improve the lives of its people, strengthen its communities and manage its resources more effectively.

This distinction is becoming even more important with the arrival of artificial intelligence.

During my involvement in broadband, digital infrastructure, smart energy and smart-city initiatives, I have seen successive waves of technology promoted as transformative. Broadband connectivity, sensors, the Internet of Things, big data and digital twins have all expanded what cities can measure and manage. Each has delivered benefits, but technology on its own has never made a city smart.

AI will be no different.

Its potential is considerable. AI can analyse information from transport networks, energy systems, water infrastructure, weather services and public facilities.

It can help cities identify patterns, anticipate problems and compare possible responses much faster than people working with fragmented systems.

Australia’s AI ambitions expose the folly of short-term broadband policy

Australia’s costly detour through copper-based NBN connections showed what happens when national infrastructure is retrofitted to suit political promises.

Traffic management is an obvious example. Rather than simply measuring congestion, an AI-assisted system could combine information about traffic flows, public transport, roadworks, major events and weather conditions. It could help operators adjust signals, redirect services or provide travellers with better information.

Similar applications are possible across the city. AI could help detect water leaks, predict infrastructure maintenance, improve waste collection, manage energy demand, monitor environmental conditions and support emergency services. In urban planning, it could allow communities and planners to explore how proposed developments might affect traffic, housing, energy use, green space and public services.

However, these possibilities should not lead us to begin with the technology. The starting point must always be the needs of the people who live and work in the city. We also discussed this last year when looking at Amsterdam.

What problems are residents experiencing? Which services are difficult to access? Where are resources being wasted? Which communities are being poorly served? Only after these questions have been answered should a city consider whether AI can help.

This is where many smart-city projects have gone wrong. They have started with a technology or a vendor proposal and then searched for a problem to which it might be applied.

Successful projects work in the opposite direction: identify the social, economic or environmental objective first and select the appropriate technology afterwards.

AI also highlights a longstanding weakness in city management: services are often organised in separate administrative and technological silos. Transport, energy, water, planning, health and emergency services may collect valuable information, but their systems do not necessarily communicate with one another.

Australia perfectly placed in critical infrastructure space race

The global space economy is becoming another layer of the world's critical infrastructure — is Australia ready to meet the challenge?

The greatest contribution of AI may therefore be its ability to assist coordination across these systems. A city should be understood as a living, interconnected community, not as a collection of separate technical projects.

That will require solid foundations. AI cannot compensate for inadequate broadband, unreliable sensors, poor-quality data or incompatible systems. Cities need open standards and interoperability so that information can be exchanged securely.

They must also avoid becoming dependent upon proprietary systems from which their own data cannot easily be retrieved.

There are equally important questions about public trust. If AI influences decisions about policing, planning, housing, transport or access to services, people are entitled to know how those decisions are made. Responsibility cannot be transferred to an algorithm.

AI should support accountable human decision-making, not replace it.

Privacy, cybersecurity, bias and surveillance must therefore be addressed from the beginning. Cities should also assess the energy consumption and environmental cost of large-scale AI systems. A technology intended to improve urban sustainability should not quietly undermine it.

Inclusion is another essential consideration. Digital services designed around technically confident users can disadvantage older people, people with disabilities, those with limited English and those without reliable digital access.

A genuinely smart city must keep offering accessible human services and ensure technological progress does not deepen existing inequality.

AI deepfakes and identity theft changing methods of proving identity

AI is destroying our old methods of proving identity and our security systems must catch up.

The most sensible approach is gradual and practical. Cities can begin with clearly defined problems, involve affected communities, test solutions on a manageable scale and publicly measure the results.

Projects that work should be designed from the outset so they can be expanded and connected with other services.

AI represents an important next stage in the evolution of smart cities. It can help us move from collecting data to understanding it, from reacting to problems to anticipating them, and from operating isolated services to coordinating the city as a whole.

But the essential principle remains unchanged: cities are built for people. AI will make our cities smarter only if it helps create places that are more liveable, inclusive, sustainable and responsive to the communities they serve.

Paul Budde is an IA columnist and managing director of independent telecommunications research and consultancy, Paul Budde Consulting. You can follow Paul on Twitter @PaulBudde.