Police face and fingerprint biometrics programs meet backlash in India
Two cities in the Indian state of Uttar Pradesh are introducing separate biometric crime-fighting technologies that are causing serious concerns regarding privacy and accuracy.
Lucknow deploys AI-powered facial expression monitoring system to curb sexual assaults
The Times of India reports that the city of Lucknow will deploy AI-enabled face biometrics to publicly monitor women in distress at 200 crime hotspots. The behavioral facial recognition technology will be integrated into the city’s public monitoring network and is part of Uttar Pradesh’s Mission Shakti initiative to prevent sexual assault against women.
According to the report, the new AI-cameras, which are stationed in areas identified as having high crime rates, will be directly connected to a dedicated police monitoring facility which will alert strategically placed patrol units to respond to suspected crimes. Authorities in Uttar Pradesh’s state capital hope that this new initiative will accelerate police response times and ultimately help lower the city’s staggering crime numbers.
However, critics of the new program fear that the AI-based public camera system raises several concerns regarding its efficacy and accuracy, writes Ivan Mehta of TNW. He fears that facial recognition tech’s low success rate in some previous initiatives as well as the software’s potential to misinterpret facial expressions will lead to misidentification and wrongful discrimination.
Aside from bias, Mehta also anticipates data privacy issues arising as it will be unclear how and where the collected facial recognition data will be stored. Further, another shortcoming of Uttar Pardesh’s surveillance project will be that it will be implemented without sufficient test data.
Mehta further writes that existing face AI lacks accuracy and reliability due to the small range of emotions its algorithms currently measure. His concerns are further backed by industry experts who voiced their concerns about AI facial expression detection in a recent interview with the MIT Technology Review. In the interview, Affectiva CEO and Co-founder Rana El Kaliouby backed these concerns and stated that a so-called “one-to-one mapping” between facial expression and emotion does not exist.
Based on this assessment, data on the true efficacy of Uttar Pardesh’s latest security initiative will most likely become available soon after the project’s implementation. In the meantime, it appears the project will meet various technical and privacy challenges that need to be addressed to ensure its overall success.
Noida to implement controversial biometric database of domestic workers
Noida, a city in the northern part of the state of Uttar Pradesh, will introduce a controversial fingerprinting program to document domestic workers’ biometrics in a bid to reduce crime, writes the Hindustan Times.
The new initiative will allow employers to fingerprint their domestic workers using a specially designed mobile application that transmits the biometric data to a Noida law enforcement database. The measure, a supplement to the city’s existing worker verification requirements, is meant to decrease theft and other related crimes allegedly committed by perpetrators who pose as domestic workers.
Police Commissioner Alok Sing stated that there is no deadline for the project’s launch. He further added, “Most of these workers are loyal to their employers. However, there are instances when some people masquerade themselves as domestic workers and commit crimes. This initiative is to check such people and take actions. We will soon announce the mobile app’s launch date and following that, the Noida Police will organise camps in different societies and ask people to register their domestic helps. We will record their fingerprints and other details which will be used for safety and security purposes.”
Yet, despite the promised increase in security, critics of the preventive measure fear that the new biometric verification measure unfairly targets innocent individuals. According to Internet Freedom Foundation Executive Director Apar Gupta, the fingerprinting application is unfair as it violates the data privacy of innocent domestic workers.