Facial recognition technology (FRT) is a form of artificial intelligence that has been growing in popularity over recent years. The uses are wide and varied, and the creative industries have embraced this tech, using it in numerous ways, whether for animation in gaming or film to make character expressions more lifelike, personalising visitors’ experiences in art galleries and museums depending on reactions to exhibits, in video games to assess emotions and engagement levels, updating digital advertising display content based on the demographics of viewers, assisting the retail sector in the prevention of crime, or monitoring employee attendance.
Like all new tech, FRT has huge potential but with opportunity comes challenge. Often public concern is around big brother type surveillance activities, however public awareness about the privacy and ethical issues is also growing, e.g. the widely reported fines for Clearview AI in a number of countries. Clearview had created its database by scraping over 20 billion facial images from across the internet, including from social media sites, breaching data protection laws.
Bias and inaccuracies can also be problematic as they can lead to discrimination, which in an employment or law enforcement context can have particularly significant consequences, e.g. the food delivery company’s FRT not recognising a driver, meaning he was unable to access work opportunities. Add to this the increasing awareness about data security, due to a number of high profile breaches, and the potential misuse of such sensitive data and you have a perfect storm (and if it goes wrong a PR nightmare!).
If you are using FRT you need to think carefully about the data protection issues that arise. The ICO has recently published guidance on biometric recognition, as well as taking enforcement action against a leisure centre operator to stop them from using FRT (and fingerprint scanning) to monitor employee attendance. You need to be clear about the lawful basis for your processing as well as ensuring you meet all your GDPR obligations. A thorough risk assessment is vital, e.g. robust security requirements are essential as once biometric data is compromised the damage is irreversible and identity theft and fraud are serious concerns. Venturing further, should AI tech be used, there is an extra layer of regulatory intricacies to navigate, such as the ban on emotional recognition within the workplace.
Fortunately, a framework of regulatory guidelines exist to help organisations navigate the FRT compliance maze and so will enable many to benefit from the versatility of FRT to enhance the creative process in a safe, secure and compliant way.