The Virus Gives AI a Chance to Prove It Can Be a Force for Good

(Bloomberg Businessweek) -- In China, doctors use artificial intelligence tools provided by Huawei Technologies Co. to detect signs of Covid-19 in CT scans. In Israel, Tyto Care Ltd. offers in-home medical exams, using AI to deliver clinical-grade data to remote doctors for diagnosis. Chinese tech giant Baidu Inc. devised an algorithm that can analyze the biological structure of the new coronavirus and made it available to scientists working on a vaccine. AI is also behind biometric identification systems being rolled out by governments to track the virus and enforce lockdown efforts, including temperature screening systems deployed throughout Beijing and CCTV cameras hooked up to facial-recognition software in Moscow. “AI is being used to fight the virus on all fronts, from screening and diagnosis to containment and drug development,” says Andy Chun, an adjunct professor at City University of Hong Kong and AI adviser at the Hong Kong Computer Science Society, a nonprofit industry group.

The pandemic is opening up a massive opportunity for the tech industry, while it shines a light on calls for more scrutiny of AI innovations being developed faster than regulators are able to devise rules to protect citizens’ rights. The quick introduction of AI tools to fight the virus is being done in the name of the greater social good, but it raises important questions around accuracy, bias, discrimination, safety, and privacy. Fever detection, facial recognition, and other forms of remote biometric identification technology can collect sensitive data, which can put people at risk if not handled properly.

The implementation of some of the technology is based on extraordinary powers used by governments to restrict their citizens’ freedoms, as well as exemptions from data protection laws. Governments, international agencies, and companies may be unwilling to part with that level of access to personal information when the crisis subsides, says Tom Fisher, a senior research officer at Privacy International, a U.K.-based nonprofit advocacy group that’s tracking tech’s role in the global response to Covid-19. “This is why we have to be critical right now, while these measures are being deployed, and make sure these measures are necessary and proportionate so we don’t get in a situation where our rights in the future are being eroded,” he says.

Even before the emergence of the coronavirus, facial recognition had become a target for privacy and civil liberties advocates, who’ve urged governments to ban the software or issue a moratorium until safeguards are put in place. People with darker skin tones and women are particularly susceptible to being misidentified: A recent global software study from the National Institute of Standards and Technology found that false positives are generally as much as 100 times more likely for Asian and black faces compared with images of white people.

Attempts to regulate AI are in their infancy. While jurisdictions including Singapore and the U.S. have released guidelines for its use, the European Union is racing to be the first to propose firm rules. In February, the EU called for feedback from citizens on its plans to regulate AI, including remote biometric identification technology, and will use that to inform proposed laws later this year. Software such as facial recognition generally can’t be used for remotely identifying people under existing EU privacy rules—with some exceptions—and the bloc wants to figure out where exactly people’s red lines lie.

U.S. and European law enforcement officials warn against bans of tools they say can make societies safer. Governments should craft careful policies instead, they argue. “In society and in police work, we are permanently confronted with errors, but suddenly with facial recognition, there’s zero tolerance,” Wim Liekens, chief information officer of the Belgian police, said in January at a privacy conference in Brussels.

The use of biometric technologies to tackle the coronavirus is part of the broader surveillance regimes of governments tracking their citizens’ compliance with restrictions on movement. In China, authorities sourced data from phone carriers and called on private companies to create AI solutions to trace all citizens’ travel patterns. In Europe, telecommunications operators are supplying governments with aggregated and anonymized mobile phone location data to monitor lockdown efforts, and some countries are pushing ahead with voluntary apps to trace whom infected people have had contact with.

Unlike China, Europe has strict privacy rules about what companies and organizations can do with people’s data, but, under special circumstances, governments can pass emergency bills to use citizen data without their consent. So far, publicly announced tracking plans in Europe have been in line with the bloc’s strict privacy rules, according to European data protection authorities. Still, the regulators say they intend to keep a watchful eye to ensure no party oversteps the bounds.

Corporations are also turning to biometric identification to protect against the virus. ASML Holding NV, a Dutch maker of semiconductor manufacturing machines, has installed an infrared thermal camera at its headquarters in Veldhoven, at the entrance to a sanitized clean room where it assembles large equipment for chipmakers. Poland’s Pragmasoft fast-tracked its remote fever-detection solution, Feverguard, in light of the crisis and already has preorders from factories and offices in Poland, with interest from potential clients in the U.S. and Serbia. ASML says it doesn’t record names and temperatures to ensure the company is in line with privacy rules. Pragmasoft says it uses low-resolution thermal sensors, obscuring any identifying physical features.

For AI developers, the coronavirus pandemic represents an opportunity to prove their tech can be a force for good, says Jon Medved, chief executive officer of Israeli private equity firm OurCrowd, which invested in Tyto, based in Netanya, Israel. “AI allows you to encounter a new reality, begin to understand, and fight back,” Medved says. “We will look back at this time and say that the Covid crisis was the coming of age of AI—not as a threat to humanity, but as a real aid in terms of fighting global dangers.” —With Ilya Khrennikov

©2020 Bloomberg L.P.