In July 2019, the UK privacy regulator, the Information Commissioner's Office (ICO) issued a warning about the privacy implications of automated facial recognition technology (AFR). The ICO was concerned that AFR "represent[s] the widespread processing of biometric data of thousands of people as they go about their daily lives."

The UK High Court recently handed down an important and timely decision in Bridges, R (on application of) v. Chief Constable of South Wales Police [2019] EWHC 2341 (Admin). The Court ruled that the South Wales Police's (SWP) use of AFR was proportional, lawful, and consistent with human rights and data protection laws. This was despite SWP's use of AFR interfering with the privacy and data protection rights of affected individuals.

This is the world's first legal challenge over the use of AFR.

Background

SWP has trialled a particular type of facial recognition technology (AFR Locate) since April 2017 with a view to it being rolled out nationally. AFR Locate works by capturing digital images from CCTV feeds of the faces of members of the public. The images are then processed in real time to extract facial biometric information. This information is compared with facial biometric information of people on police watch lists. If there is no match between the images, the data is immediately deleted after being processed.

The legal challenge was brought by the human rights campaign group Liberty on behalf of a Cardiff resident, Ed Bridges. Mr. Bridges argued that SWP's use of AFR contravened his human rights as well as data protection laws.

The question for the Court was whether the current legal regime is adequate to ensure appropriate and non-arbitrary use of AFR.

Key findings

The High Court dismissed Mr. Bridges' claim on all grounds. Leaving aside claims relating to the breaches of human rights laws, the key data protection findings were as follows.

  • Justifiable processing: Biometric data captured by AFR is the personal data of people who are not on a police watch list. The Court considered that members of the public whose images are captured by AFR are sufficiently individuated from all others. Although this processing of the biometric data was also "sensitive processing," the Court ruled that it was justified. For SWP to achieve its purpose of identifying persons on watch lists, biometric information must be processed in the first place. The processing of this data is necessary for SWP's legitimate interests to detect and prevent crime.
  • Law enforcement processing: Biometric data processing does not contravene the data protection principle that any personal data processing for law enforcement purposes must be lawful and fair.
  • Data protection impact assessment (DPIA): SWP's DPIA for AFR complied with UK law. The Court ruled that SWP's DPIA set out a clear narrative that took account of the potential breaches of UK human rights laws. It also identified safeguards that are in place to determine what personal data will be retained and why.
  • Appropriate use: The current legal regime is adequate to ensure appropriate and non-arbitrary use of AFR. SWP's use of AFR Locate is consistent with human rights and data protection laws.

Comment

It has always been difficult to achieve a regulatory balance between harnessing new technologies and safeguarding the privacy and data protection rights of affected individuals. This ruling is fact-specific and should not be interpreted as a UK-wide green light to use AFR. However, the ruling nonetheless provides much needed judicial clarity on AFR. The ICO, which has been critical of police and private use of facial recognition technology, has said it will review the judgment carefully. As such, we expect the ICO to issue further recommendations and guidance to law enforcement about how to deploy any future AFR technology. Keep an eye on this blog as we'll be sure to keep you fully up-to-date.

The content of this article is intended to provide a general guide to the subject matter. Specialist advice should be sought about your specific circumstances.