The Indian government may soon bring in an Aadhaar-based facial recognition software to replace biometric fingerprint or iris scan machines at Covid-19 vaccination centres across the country.
The announcement was made by National Health Authority CEO RS Sharma in an interview with ThePrint. Sharma cited an ongoing pilot project in Jharkhand, where he claimed the government had been conducting 1,000 authentications through facial recognition on a daily basis.
Sharma has said the deployment of the facial recognition software would ensure a ‘touchless’ Covid-19 vaccination process. Presently, beneficiaries are required to touch the fingerprint scan machines for identity verification, before getting their vaccine jab, raising fears about the spread of Covid-19 at vaccination centres.
“Imagine a person who generated their Aadhaar card in 2011. Even after a decade, the software is able to recognise the face. Once we do about 50,000 to 60,000 facial authentications under the pilot, we will roll it out across the country,” Sharma told the publication.
The government is bringing in FRT after reports of failed Aadhar-based fingerprint authentication depriving genuine beneficiaries of welfare benefits have been doing the rounds for a few years now. However, large-scale implementation of facial recognition tech (FRT) by private companies and the Indian government isn’t without legal scrutiny or privacy pitfalls.
Is Facial Recognition Tech Secure?
In light of the use of facial recognition, an Indian digital liberties organisation, the Internet Freedom Foundation (IFF) has also filed a right-to-information or RTI request with the government about its plans of implementing its FRT for the Covid-19 vaccination drive. The IFF has claimed that even globally, no FRT has been found to have a 100% success rate and that implementation of an error-prone system without adequate legislation containing mandatory safeguards, would deprive citizens of essential services, while also raising concerns about privacy.
The organisation has cited the example of the FRT adopted by Delhi Police, which in 2018 had an accuracy rate of around 2%, and which subsequently fell to 1% in 2019. Besides, India doesn’t have specific laws governing the use of FRT, while the Personal Data Protection Bill, 2019, is yet to be passed.
There are more questions about the efficacy of the technology.
Recently, the case of an Uber driver named Srikanth came to light. Srikanth shaved his head while on a religious pilgrimage. But when he came back, Uber’s FRT disallowed him from accessing the app, depriving him of his only source of income, as alleged by the Indian Federation of App-Based Transport Workers (IFAT), a union of gig and platform workers. Uber, on the other hand, has claimed that its FRT is able to recognise natural changes in a person’s appearance, also alleging that Srikanth had been blocked because he was in violation of the company’s ‘community guidelines’. The company hasn’t disclosed the exact provision of the community guidelines that Srikanth allegedly violated.