Is Facial Recognition Stooping Down with Unauthorized Surveillance?

Agencies, both public and private should have a better surveillance policy with facial recognition systems

How often do you think your face appears on the internet? Well, if you are not popular enough, then maybe very rarely right? Although, your face might appear in many databases of facial recognition technology, even without consent. Then you may ask, isn't this invasion of privacy? Absolutely yes.

Facial recognition is reigning the digital identity verification system because of its uniqueness and efficiency to identify faces even from vague footage. Facial recognition is considered one of the emerging biometric systems that can enhance security and the digital payment process. But, there are many controversies and concerns looming over facial recognition technology, especially how the state governments and federal agencies are using it.

Are We Losing It?

While facial recognition systems promise to improve public security systems, it also tends to misidentify people a lot. According to a New York Times article, Nijeer Park, a New Jersey man was wrongfully accused of shoplifting and trying to hit an officer with a car. Further, the article says that he is the third black man to be wrongfully arrested based on facial recognition. Facial recognition was found to be biased towards people of colours and genders other than male.

Monitoring people through surveillance systems can be a bit intrusive since they do not work on consent every time. The collected database of faces can be shared among various interfaces without our knowledge. These are serious hampering of privacy and hence, needs to be addressed.

A recent study published by Inioluwah Deborah Raji and Genevieve Fried surveys over 100 face datasets between 1976 to 2019 of 145 million images from different sources and demographics. The study reveals that the advent of deep learning demanded a larger pool of data thus making it uncontrollable and unmanageable to obtain subject consent. Further, the report states the role of the DeepFace model introduced in 2014 by Facebook in gathering the largest facial datasets from Facebook profiles and commercializing the technology.

The MIT Technology Review article quotes Deborah Raji who hopes that "the paper will provoke researchers to reflect on the trade-off between the performance gains derived from deep learning and the loss of consent, meticulous data verification, and thorough documentation."

There needs to be strict regulations and policies against the commercialized usage of facial databases by both private and government agencies without the consent of the person. For example, the Illinois Biometric Information Privacy Act says that companies should notify the public and obtain written consent from the individuals in case they want to possess their biometric details.

The growing concerns over unregulated surveillance should be addressed at the earliest to define privacy protection in emerging technologies like facial recognition. An ethical approach and transparent communication from the federal, state, and private entities will enable better monitoring by respecting an individual's right to privacy.