The use of facial recognition devices has been increasing at an unprecedented rate. It’s used for surveillance everywhere, from malls to airports to law enforcement. While this technology comes with some upsides, its exploitative features cause concerns about its users' safety and privacy.

1.    Data Privacy

The data received from the face recognition device contains numerous video and image files. It also includes the maps that are created with facial recognition systems.

All these files are stored on servers and companies can access them through a cloud. Like usual computer systems, this data, too, can be accessed by hackers.

Many smartphone applications and companies use this data for research. Consequently, they have a chance to garner profits out of it — without you knowing about it. These companies can also share or sell this data to other third parties who can track and recognize you.

Companies that develop face recognition apps must take data privacy into account. The best way to do this is to write a comprehensive privacy policy and get the customer’s consent.

2.    Racial Bias

Facial recognition features display racial bias. Reports have shown that facial recognition devices fall short when identifying women and people of color.

Facial recognition devices need to be fed with diverse datasets to learn how they can identify a diverse group of people with different characteristics. Suppose tech companies want to perfect facial recognition systems. In that case, they must feed a diverse dataset of people belonging to other races, genders, and skin colors, and not just a particularly privileged group.

3.    Low Reliability

According to a research study published by MIT, misidentification in facial recognition devices is rampant.

Currently, the factors like poor image and video quality and low illumination can throw off the system making it unable and unreliable to identify a person.

Moreover, if a person is standing at a slight angle, have worn glasses, or have gotten a haircut, the facial recognition device may fail to recognize. These limitations present severe flaws in the system.

4.    Lack of Regulation

Currently, there’s no regulatory framework or legislation around facial recognition devices and data uses. The state of Illinois in the US passed legislation for biometric identification. Moreover, many cities across the world have banned the use of facial recognition by the government altogether.

Businesses that are adopting facial recognition must plan the risks to privacy and safety that may arise in the future. While it’s challenging, a robust cybersecurity infrastructure can ensure that it works well. 

Lean Security is an online security services provider focusing on providing managed security services  in Gordon, NSW. Our services include web and mobile application penetration testing, API and IoT penetration testing, and web security audit. Contact us at+61-2-8078-6952 in case to learn more.