Are Cities Being Smart About Artificial Intelligence?

The complexity of creating a smart city is growing, compounded by the 1 billion gigabytes of data per day that is generated from connected (IoT) devices. Government leaders report being increasingly stressed about how to manage, manipulate and make sense from the massive amount of information being collected and exchanged on everything from pedestrian movement to basic urban operations.

Artificial intelligence (AI), or “the ability for a machine to perform cognitive functions as well as or better than humans” as defined by the Future Today Institute, may hold the key for the future of city making. AI represents the third era of computing, generally defined as training machines on activities such as perception, learning, reasoning, problem-solving, contextual understanding, making inferences, predictions, and exercising creativity.

In the next decade, AI will reshape our world by automating the mundane and enhancing the knowledge economy, speeding up discovery and innovation so that we can finally understand the information on the world around us. That has a big impact on the future of how we design urban systems. AI can unlock the power of data and translate the bits and bytes so that ready humans can turn that into insights and action. 

But do we want machines to be that smart?

Most everyone is somewhat familiar with the dystopian, fear-based narrative that AI will replace humanity in the most sinister of ways. Beyond making a good movie, this story line provides a cautionary tale that deserves serious attention and consideration. There are many reasons, from algorithm bias to data privacy issues, to be careful and diligent about how we program machines to think for us. 

There is an urgency to be intentional and unrelenting in our desire to shape technology for the betterment of society. If designed correctly, AI may be able to free humanity and we can finally create the communities and experiences that we crave. If we can find a way for the machines to do what machines do best, then humans can do what we do best – which is to be creative, compassionate, innovative, resourceful and to unleash the power of love. No machine can do that. 

But what about the private sector and their motives? 

AI is already omnipresent. From providing faster access to our passwords to intuitive product marketing campaigns that border on spooky, most every person has some engagement with AI, even if they don’t realize it. The private sector has invested in advancing the technology, motivated by a financial incentive to monetize data. Society has proven a willingness to give up privacy in exchange for convenience. 

But when this translates to the public sector, the rules change. Government must serve all citizens, residents and visitors, and therefore the need for communication, transparency and trust building is essential. It is incumbent that we build AI systems with goals that align with society’s values. But first, we must provide the frameworks that help people explore and express those values. 

We need to design systems that are inclusive and accommodate different perspectives while providing leadership that doesn’t accommodate divisive practices. Those systems need to be nimble, adapt to a quickly and constantly changing reality. The private sector has vast resources to contribute, but it is up to the broader community to define the guard rails for how we use those technologies. That only happens when there is a multi-sector conversation. We have to do a better job of providing space and agency to listen to each other – to find commonly held beliefs before we default to divisive tropes. 

When developing AI technologies and systems, what kinds of questions should we ask? 

Often when we talk about technologies, we get mired down in the what and the how. What technologies should be applied? How should we design policies to either enable or regulate these technologies? What is the role of government, what is the role of industry? How will we encourage innovation while also keeping in mind the interest of the public good? How can policies and technologies be designed to be equitable and inclusive? 

Perhaps the more important questions that precede the ‘what’ and the ‘how’ could be the ‘who’ and the ‘why’. Who is currently driving these conversations? Why are they engaged? Who is NOT included or even invited to be a part of the discussion? Why are they not included? Exploring these questions gets to the heart of what later drives the what and the how. It can be a very disruptive action to say “why are we anticipating solutions for people who are not in the room? Why are they not here – why are their voices not represented?”  Smart city leaders in every sector will be well served to spend more time on the why and the why which can add some much-needed clarity to the what and the how. We have so much to learn from each other if we just listen.

So, what’s next for AI and smart cities?

The ultimate hope of smart cities is to be able to ethically and responsibly use the power of data processing, analytics and AI enabled by 5G to unlock data in order to inspire a better collective understanding of how to improve our cities. A shared understanding can increase public engagement and invite new experiences and perspectives to create unique ideas and solutions.

Achieving this means inviting new levels of collaboration. It is not up to one person. It is not up to one sector. We will need to re-imagine cooperation. This will be very challenging and will stretch us in uncomfortable ways. But it is possible and much needed. If we can do this, perhaps we can rebuild our communities’ human connection. Perhaps we can restore faith and trust in each other, in our systems, in our government. 

It is important to realize that this utopian vision is a huge technical feat – we will need advanced technology, especially AI, to parse and quickly analyze the data in meaningful ways that we can all understand and relate to. Humans will simply take too long. And we can’t wait.