**4. Conclusions**

### *4.1. Are Artificially Intelligent Cities on the Horizon?*

According to Andrew Ng, cofounder of Google Brain, "AI is the new electricity. Just as 100 years ago electricity transformed industry after industry, AI will now do the same." The impact of AI will go beyond the industry; it is set to change the world [111]. An internationally conducted survey [112] highlighted that "the prospect of an AI future both excites and concerns people around the globe. Nonetheless, fears around the drawbacks of AI are offset by the benefits, and the net result is positive. AI will likely to change society for the better." AI applications have also significant potential to transform our cities. This may lead to the next-generation smart cities [113] being coined as "artificially intelligent cities". Building artificially intelligent cities may save our civilization from the earlier mentioned catastrophes, but it all depends on how we design and use AI, and on who will profit from it [114]. The risk here is for AI to become a vehicle for increasing the wealth of the top 1% of income earners (i.e., top 10 wealthiest people in the world and monopolistic multinational corporations) and the power of biased and unethical politicians [115].

Time will tell if AI systems make our cities "smart enough" to provide better living conditions for all (i.e., people, flora, and fauna coexisting in urban ecosystems). To date, while there have been significant technological advances, these have not been matched with innovations in governance mechanisms. In addition, the policy apparatuses of most local governments need significant modernization to take full advantage of technology affordances in an agile manner [116].

If there is one thing that cities and local governments have learned from the ongoing COVID-19 pandemic, caused by the SARS-CoV-2 virus, is that when they are willing, they can respond in a proactive and agile manner to the changing environmental conditions. We hope that cities will keep on this track after the current crises pass, modernizing their governance mechanisms and policy frameworks to take full advantage of emerging technologies—particularly AI.

The COVID-19 pandemic has also demonstrated that local governments need to seriously consider their digital infrastructure capabilities and capacities. For example, when the Queensland state governmen<sup>t</sup> (in Australia) decided to make education available online, its infrastructure failed to deliver the public service (i.e., provision of online education) due to significant web traffic. A few schools had backups in place with paper resources but the problem highlighted significant issues with existing networks [117]. Online education has also highlighted social equity issues associated with the digital divide with some lower income students struggling to meet the required technology capabilities [118], and special needs students, including those who speak a language other than English, struggling to receive the required one-on-one assistance [119].

In this paper, we mainly focused on the artificial narrow intelligence level of AI. Nevertheless, if somehow one day we manage to build artificial general intelligence or artificial superintelligence (these two AI levels also correspond to singularity, that is, in simple terms, the intelligence explosion), we need to do all it takes for it to be, as Tegmark [120] calls it, a "Friendly-AI" (a superintelligence

whose goals are aligned with ours). Speculation on how to build artificial general intelligence or artificial superintelligence or singularity that would reshape our cities, societies, and civilization is beyond the scope of this viewpoint.

### *4.2. What Are the Key Lines of Research Concerning Artificial Intelligence and Cities?*

There are some important issues, in the context of AI and cities, that prospective research must address in order to provide our cities and societies with the best technological outcomes. We strongly believe that further investigating some of the critical issues in prospective research projects by scholars of this highly interdisciplinary field will shed light on the better conceptualization and practice of AI (artificial narrow intelligence level) in the context of cities and societies. These issues are listed below:


Considering AI's current ability to ingest big data for exploratory studies and real-time decisionmaking, it would be worthwhile to address the following research question (in addition to the above list): How can AI be used to find what we may have missed in terms of developing better (e.g., fairer and more productive) social structures, social geography, social good, political structures, economic structures, energy sources, modes of transportation, design of living structures and spaces (i.e., in normal and disaster times, such as those that COVID-19 and similar pandemics could bring on us), and so on?

The concept of AI advising us on human sociology and similar matters may sound very o ffensive to some, but when properly done, AI is merely a tool that can be used by humans for their advantage (in the sense of artificial narrow intelligence). Humans tend to learn and incrementally apply the acquired knowledge into practice. AI can analyze ideas for us faster and more in depth, and together with other developments in technologies (e.g., AR, high-performance computing, IoT, and big data), it could allow us to study and predict the potential harms and benefits of alternative ideologies, and develop better futures. Moreover, AI, in the context of artificially intelligent cities, can also help us transform our cities into smarter and more prosperous and creative ones [121–123].

Lastly, we conclude the paper by elaborating on the question we raised in the title of this paper—Can building artificially intelligent cities safeguard humanity from natural disasters, pandemics, and other catastrophes? The existing AI literature reviewed in this paper, unfortunately, does not allow us to answer it with a confident "yes". The answer to this question depends on the findings of the studies focusing on the above-listed critical questions. While we continue to have hope that AI technology will help fix or at least ease the problems created by us, perhaps another important issue is whether we will be able to use AI for the common good of all—rather than the so-called 1% [124] that is already in control of the world economy.

On that very point, at his Turing Lecture on deep learning for AI, Yoshua Bengio [125] highlighted the critical importance of using AI for social good and introduced two actionable items: (a) favoring machine learning applications to help the poorest countries fight climate change, improve healthcare and education, and so on; and (b) forming the concept of AI commons and coordinate, prioritize, and channel funding for the use of AI for social good. As stated by Hager et al. [126], "AI can be a major force for social good; but it depends on how we shape this new technology, and the questions we use to inspire young researchers."

**Author Contributions:** T.Y. designed and supervised the study, and finalized the manuscript. K.C.D., R.M. and J.M.C. contributed to the write-up of the manuscript, and improved the rigor, relevance, and reach of the study. L.B. and E.W. prepared the first draft of the manuscript and assisted in data collection. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Acknowledgments:** This research did not receive any specific gran<sup>t</sup> from funding agencies in the public, commercial, or not-for-profit sectors. The authors thank the managing editor and three anonymous referees for their invaluable comments on an earlier version of the manuscript.

**Conflicts of Interest:** The authors declare no conflict of interest.
