Need India-specific data to properly implement facial recognition tech: Infosys co-founder
As India plans to roll out a nationwide facial recognition system this year, Infosys Co-founder Kris Gopalakrishnan believes that the country must develop its own databases for efficient implementation of breakthrough technologies that use Artificial Intelligence (AI) and Machine Learning (ML).
A facial recognition system is a technology capable of identifying or verifying a person by analysing patterns based on the person's facial textures and shape.
Gopalakrishnan noted that India should carry out its own trials before implementing the facial recognition systems, as the algorithms used to train these mostly employ data of white men belonging to the Anglo-Saxon community, and it is unclear whether it will work properly in the country.
"We also need to look at biases. One of the reasons why I believe India must do research in AI and ML particularly is because most of the databases that are used to train these systems which we use today are being trained with data which is not from India," he said in an interview on the sidelines of the Infosys Prize ceremony.
Gopalakrishnan, who is also the trustee at Infosys Science Foundation (ISF), noted that AI is a tool which impacts every industry, including research.
The Infosys co-founder explained that AI and ML can be used to further leverage data to create new products, and new services like facial recognition, text translation, and diagnostics.
Facial recognition is typically used as access control in security systems, and can be compared to other biometrics such as fingerprint or eye iris recognition systems.
The National Crime Records Bureau (NCRB) last year said the proposed facial recognition system will help catch criminals, and find missing children.
Asked about the concerns raised by activists and tech experts who warn of the risks to privacy and from increased surveillance, Gopalkrishnan said "We need to sit down and come out with ethics governing the facial recognition as it is an issue that the people all over the world are grappling with."
"We need to start building our own databases. Training databases must include sufficient data from India. It is a chicken and egg problem. Because when you have huge amount of data, if you want to leverage the computing power to analyse that data, and come out with patterns or inferences, machine learning is the tool to use to find the patterns for you," he said.
Gopalakrishnan added that a soft stance may be required to train Indians to use these systems, understand the implications, as well as the downside and go forward.
"There is a lot of debate about using facial recognition in even unlocking your phone, but now they are going back to biometrics or iris recognition because in some way the facial recognition didn't work as it should be," he said.
"But we should be doing this work. Don't wait for somebody else's tool to come to India. We should start collecting our own data that is is very important," he added.
Citing another example of AI and ML use, Gopalakrishnan said scientists can use standardised DNA sequencing databases, specific to India, to develop new molecules and treatment protocols, with these systems trained using this native data.
"We should be doing our own trials before deciding whether it works in India or not. The human being is as important as these tools because we make the decision to either not use the data, or to fine tune the data," Gopalakrishnan added.
(Edited by Saheli Sen Gupta)