Smart Cities in the Time of COVID-19

Summary of Updates

In recent months, communities across the globe have experimented with adapting smart city technologies in response to the outbreak of COVID-19. Digital solutions created to better manage public safety, transportation, energy use, and other aspects of urban infrastructure are now being used for epidemiological surveillance, public health communication, strategic planning, and policy monitoring and evaluation with the aim of mitigating the effects of the ongoing pandemic.

Given the virus’s long incubation period, digital contact tracing has emerged as a method of tracking, and thus slowing, its spread. Digital contact tracing — distinct from, and potentially complementary to, human-driven contact tracing by questionnaire — has been a component of national public health responses, with varying outcomes, in Singapore, South Korea, and India. Oxford Business Group called South Korea’s Smart City Data Hub “one of the most successful demonstrations of the power of smart city technologies” by virtue of its role in “rapidly reduc[ing] infection rates without a full lockdown.”

National and local authorities worldwide have also been exploring the feasibility of using drones in the fight against COVID-19. In China, Spain, France, Italy, and Western Australia, police forces launched drones to issue public service announcements via loudspeaker urging individuals to adhere to public health regulations. Public health authorities in Oman, Saudi Arabia, Colombia, and Connecticut have experimented with using drones to detect high body temperatures that could be symptomatic of COVID-19. Drones were also reportedly used to spray disinfectant in public spaces in China, India, and Indonesia. Meanwhile, Draganfly Inc. partnered with a police department in Westport, Connecticut to conduct a social distancing measurement test program, using drone technology to gather real-time “anonymized” data on social distancing practices. However, the validity of each of these approaches has been questioned due to performance accuracy issues, additional health hazards posed, or public concern over privacy issues.

Smart city solutions have also been adapted for public health planning, monitoring, and evaluation purposes by leveraging data from multiple sources within an urban environment. In March, 45 Indian smart city command and control centers were transformed into the nerve centers for operations to contain the spread of the virus. In the same month, the United Arab Emirates had a “laboratory testing facility” for COVID-19 built in Abu Dhabi’s Masdar City, the Financial Times reported. Furthermore, nearly 30 United Nations entities have leveraged innovative approaches to respond to the pandemic, with many launching online dashboards with real-time information on COVID-19 cases and related topics ranging from food security to the labor market.

One New Scientific Discovery

The Newcastle University Urban Observatory, which “makes use of thousands of sensors and data sharing agreements” to monitor movement, “energy consumption, air quality, climate, and many other variables” in Newcastle upon Tyne, England, has been using smart city technology to study social distancing. They developed “algorithms that can automatically measure social distancing in public areas,” making it “possible to identify bottlenecks where social distancing cannot be maintained, and [determine] how citizens adapt as restrictions are imposed or lifted.”

Three Companies/Products to Watch

Draganfly Inc., a Canadian company that has been in the unmanned aerial vehicle industry since the 1990s, has launched a global dealer network that includes US and Canadian dealers and has attracted “interest from entities in Latin America, Europe, Africa, the Middle East, and Asia.”

Mitie Security, a UK-based facilities management company, has launched a suite of fever-screening solutions that detect the temperature of individuals as they enter public spaces, hospitals, retail locations, and other buildings.

Technology companies Cradlepoint and Rigado are “join[ing] forces to provide a flexible IoT-based safe workplace solution” to help US organizations and businesses get back into their traditional workspaces while adhering to US Center for Disease Control and Prevention (CDC) safety guidelines.

Three Fundings/Approvals

Amid debate on political risks and privacy concerns, Alabama, South Carolina, and North Dakota have committed to using Apple and Google’s application programming interface (API) to create Bluetooth-enabled “exposure notification” applications.

The drone company Zipline, which first launched a national drone-delivery service in partnership with the Government of Rwanda, recently made its US debut “ferrying COVID-19 supplies to hospitals in North Carolina run by Novant Health,” according to Axios.

Scandit, a Zurich-based international mobile data capture technology company, has raised $80 million in Series C funding as the pandemic has increased demand for “contactless solutions,” including “app-based [barcode] scanning, text recognition, object recognition and [augmented reality] solutions that work on camera-equipped mobile devices,” according to Mobile Marketer.

Challenges to Consider

Many of the challenges inherent in the use of smart city technologies to counter COVID-19 have been illustrated in the experimentation with drones. The deployment of drone technology in “a range of applications during the pandemic” has raised the need for “more critical thinking” about how they can add value in this context, WeRobotics Executive Director Patrick Meier argued in a World Economic Forum article. The same could be said of all of the smart city technologies that have been presented as digital solutions to the pandemic.

Perhaps the most fundamental challenge inherent in the adoption of a smart cities technology to counter COVID-19 is to ensure that the digital “solution” being presented does in fact help, rather than harm, public health. The use of drones to spray disinfectant, which may have posed more health hazards than it mitigated, is a case in point.

An additional challenge pertains to the performance accuracy of artificially intelligent technologies. If a technology, when tested, cannot meet a certain threshold of accuracy in performing its intended function, its developers may opt — or be instructed — continue working on the technology instead of deploying it. Such was the case of the Omani temperature-taking drones, which reportedly faced difficulty recording the temperature of people in an outside environment. However, virtually no advanced machine-learning technology is exempt from some margin of error. Ultimately, the choice of just how much — and what kind — of error is to be considered “acceptable” is, in the context of public health, a public policy question.

In an argument suggesting that accuracy is liable to be sacrificed in favor of urgency, University of Melbourne researchers Dr. Niels Wouters and Dr. Ryan Kelly warned against the use of facial recognition tools to detect COVID-19 in their article “The danger of surveillance tech post COVID-19.” False positives are likely to occur, they argue, especially if the datasets that these tools are trained on are “too small” or “haphazardly developed.” (They give the potential example of someone falsely classified as feverish who has an “elevated temperature” not from COVID-19 but rather from overexertion during “a weekly grocery run”). Furthermore, the stakes of being falsely labeled COVID-19-positive may be much higher in some countries than others. In democratic countries, such individuals may simply undergo “compulsory self-quarantine,” but elsewhere, governments “may decide to publicly shame” them “or limit their freedom to move,” they posit.

In an argument suggesting that accuracy is liable to be sacrificed in favor of urgency, University of Melbourne researchers Dr. Niels Wouters and Dr. Ryan Kelly warn against the use of facial recognition tools to detect COVID-19 in their article “The danger of surveillance tech post COVID-19.” False positives are likely to occur, they argue, especially if the datasets that these tools are trained on are “too small” or “haphazardly developed.” (They give the potential example of someone falsely classified as feverish who has an “elevated temperature” not from COVID-19 but rather from overexertion during “a weekly grocery run”). Furthermore, the stakes of being falsely labeled COVID-19-positive may be much higher in some countries than others. In democratic countries, such individuals may simply undergo “compulsory self-quarantine,” but elsewhere, governments “may decide to publicly shame” them “or limit their freedom to move,” they posit.

Classic privacy concerns have also surfaced as a significant challenge surrounding the use of facial recognition, contact-tracing, and other technologies that involve the collection of personal data. Wouters and Kelly argue that contact-tracing applications raise “unanswered legal, constitutional, and democratic questions.” In an article published by the Brookings Institution, Professors Julie E. Cohen, Woodrow Hartzog, and Laura Moy warn, “Contact tracing done wrong threatens privacy and invites mission creep into adjacent fields, including policing.” They note that the business models of the private-sector technology platforms adopted by public health departments “magnify the risks of data leakage” despite assuring users in their privacy policies that no data sharing will occur. Two such instances of data leakage have already occurred with North and South Dakota’s Care19 app and China’s Alipay Health Code, according to the article.

The authors of the Brookings Institution article raise an additional, overarching concern: the risks entailed in ceding authority in the pandemic response to private-sector technology firms. They argue that “uncritical deference” to private-sector solutions is “counterproductive” because it lets “insufficiently regulated platforms” define “key values,” “deepens preexisting patterns of inequality in society,” and “ignores” the technology industry’s “role in driving public polarization and amplifying the misinformation and disinformation that undermine public-health efforts.” Instead, they propose “collaboration around human-centered public-health surveillance” with “strong data protection mandated by law and extending into emergency response systems,” as well as “deeper, more structural reforms regulating the platforms themselves.”