Why human collaboration is the key to unleashing AI’s potential

If anyone required convincing of the need to make our cities smarter, more efficient, and increasingly sustainable, just consider that the global population is expected to hit nine billion people by 2037, and a majority of these people will live in cities and large metro areas. In fact, it is estimated that 70% of people will live in cities by the second half of the 21st century.

This global population explosion is just one of the factors that motivates me to work on AI-based solutions for smart cities as a professor at Mohamed bin Zayed University of Artificial Intelligence (MBZUAI). The concept of the smart city — in which advanced technologies and data analytics are used to improve the quality of life, enhance sustainability, and optimise services such as transportation, energy, and public safety — is not new.

But with the rise of AI, it is set to take a giant leap forward, and this progression will be essential to the smooth functioning of cities, and indeed of society itself, as the population moves inexorably towards nine billion people.

From cars to smart metres

AI, it could be argued, is the latest in a long list of technologies including cloud computing, 5G and edge, that will enable smart cities to become a reality. AI can analyse vast amounts of data gathered from numerous sources including traffic management, security cameras, public utilities, and hospitals to identify patterns and help identify better ways of doing things.

Think of a future with fewer traffic snarl-ups as autonomous vehicles select the optimal routes for their journeys, smart grids that mean all energy produced is used or distributed elsewhere, and an abundance of fresh food produced locally in vertical farms assisted by AI.

But let’s not assume the journey will be easy, as there are challenges to achieving these goals. The pace of smart city development remains somewhat stymied by a lack of data and knowledge-sharing by the organisations that make up cities, while the rapid proliferation of connected devices — from cars to smart metres — raises the threat of cyberattacks.

First, let’s look at the need for data, which is a key requirementfor the development of new AI-based services. It is a sad irony that, while data is increasing exponentially and is projected to grow to more than 180 zettabytes globally by 2025, it often remains difficult for researchers and developers to acquire it for R&D for work on smart city innovations.

Part of this challenge stems from the sensitivity of data and the need to protect customer privacy, and the fact that many organisations such as utilities and have not historically needed to share significant amounts of data. Thankfully, these issues can be addressed by anonymising and generalising data, and by embracing a ‘federated learning’ model which means data can be analysed on-site where it was captured rather than being sent to another location.

Smart city success

But one of the most vital elements necessary for smart city success remains coordination, the need for the people who lead, run, and manage cities to collaborate more, especially where AI is concerned. For cities to become truly smart, the traditionally siloed services of cities need to become more interconnected, allowing for greater collaboration, knowledge sharing, and cross pollination of ideas.

This notion of citywide executive coordination is just one of the many benefits of MBZUAI’s executive AI program (MEP), which brings together senior leaders from top public and private sector organisations to learn about AI and its potential to transform their industries. 83 executives, mostly from the UAE, graduated from the second and third cohorts of the program in March.

I was fascinated to see the teamwork and collaboration between the CEOs, directors general, and executive directors as they got to know each other and teamed-up to work on capstone projectsto apply their new found AI learning to real-world challenges inareas including health care, security, environment, education, and energy.

By focusing on new challenges and learning about each other’s industries, the executives gained a different perspective on their own sectors, and of course, the value of data. One of the longer-term benefits for the UAE and the wider region, I believe, will be greater smart city collaboration between and among organisations.

It was the American historian and sociologist Lewis Mumford Lewis who described the city as “a vast and intricate organism, a living thing whose parts and organs must work together to maintain the whole.”

As AI technology advances and brings us closer to achieving smarter, more sustainable cities, it’s worth reminding ourselves of the importance of human collaboration to ensure all the dots are connected.

Mohsen Guizani is a Professor of Machine Learning at Mohamed bin Zayed University of Artificial Intelligence