As India prepares to install a nationwide facial recognition system in an effort to catch criminals and locate missing children, human rights and technology experts on Thursday warned of the risks of privacy and increased surveillance.
According to the National Crime Bureau of India, the use of camera technology is an attempt to “modernize the police force, gather information, identify criminals, and verify”.
One of the world’s largest facial recognition systems, the official contract will be awarded on Friday.
But little is known about where it will be set up, what data will be used for and how data storage will be regulated, said Apar Gupta, executive director of the non-profit Internet Freedom Foundation.
He told the Thomson Reuters Foundation: “It’s a mass surveillance system that collects information in public places without any underlying reason.”
“Without a data protection law and an electronic surveillance framework, this could lead to social policing and control,” he said.
An Indian Home Ministry spokesman declined to comment.
Globally, the rise of cloud computing and artificial intelligence technology has popularized the use of face recognition for a variety of applications, from tracking criminals to catching the wrong students.
There has been a growing response, however, and San Francisco authorities have banned the use of facial recognition technology by city workers, and “anti-surveillance fashion” is becoming popular.
Face recognition technology was introduced at several Indian airports in July, and Delhi police said last year that they had identified about 3,000 missing children in just a few days during a trial.
But technology site Comparitech, which in a recent report ranked India’s Delhi and Chennai among the world’s most surveillance cities, said it had found “a slight correlation between the number of universal CCTV cameras and crime or security.”
Indian authorities say face recognition technology is needed to strengthen a severely under-policed country.
According to the United Nations, there are 144 police officers for every 100,000 citizens in the world.
The technology has been shown to be inaccurate in identifying black women, ethnic minorities and transgender people.
Thus, its use in a criminal justice system where vulnerable groups, such as indigenous peoples and minorities, are over-represented carries a greater risk of abuse, says Bidushi Marda, a lawyer and artificial intelligence researcher at Article 19, a UK-based human rights organization.
“The use of facial recognition provides a gap between technological objectivity and institutionalizing systemic inequalities without delivering its promise,” he said.
“Observing security will become synonymous, simply because of a constant, perpetual curfew on individual autonomy. It creates the risk of further alignment and inequality of the weaker parts.”
The Supreme Court of India, in a landmark judgment in 2017 regarding the National Biometric Identity Card Program Aadhaar, stated that privacy is a fundamental right, amid concerns over data breaches and the mandatory use of the card for services.
However, the verdict did not check the rollout of facial recognition technology or the proposal to link Aadhaar with social media accounts, Gupta said.
“There is a significant increase in national security as a central basis of policy design. But that cannot be a reason to limit national security rights,” he said.
“It is extremely worrying that technology is being used by the state as an instrument of power instead of a tool for empowering citizens.”
ম Thomson Reuters 2019