A new audit system, created by Minderoo Centre for Technology and Democracy, was used to assess the Police’s compliance with standards of transparency, equality, and fairness surrounding their use of AI facial recognition.  

Using UK guidelines in the Data Protection and Equality acts, as well as previous court cases, the audit assessed the use of facial recognition in three case studies in England and Wales. 

“Building a unique audit system enabled us to examine the issues of privacy, equality, accountability, and oversight that should accompany any use of such technologies by the police,” Professor Gina Neff, Executive Director at the Minderoo Centre for Technology and Democracy said.

Case Studies in the UK

The case studies included the Bridges court case in South Wales, where Police used Facial Recognition Technology (FRT) to scan crowds and compare faces to known criminals in real time. The Court of Appeals found the use of FRT unlawful. 

In another case, the Met’s use of live FRT would likely have been found unlawful if challenged in court as there was a lack of oversight, transparency and formal review in how the technology was used. 

The Met Police’s use of FRT also did not have any demographic reviews to evaluate if it perpetuated racial profiling. Internal reviews into the technology’s racial and gender bias were also not disclosed to the public. 

A further case study in the South Wales department was investigated, this time in their use of mobile phone operated facial recognition. While the report noted improvements, a major concern was the retention of images of innocent people who were arrested but not convicted, which are unlawful to keep. 

Findings 

All three case studies lead to the same conclusions: a lack of transparency, accountability and oversight plagued the use of facial recognition in the police force. 

The Police have largely hidden the amount of data that is harvested and have not published their internal audits to find if their own technology is biassed, the report found. 

A lack of formal legislation regulating the use of AI and facial recognition leaves little room for accountability for when these technologies are used by police. 

Evani Radiya-Dixit, a visiting fellow at Cambridge’s Minderoo Centre and lead author, noted the “lack of robust redress mechanisms” available to the public to legally challenge the use of facial recognition by police.

The House of Lords Committee on Justice and Home Affairs noted the lack of formal legislation regulating the use of facial recognition, leaving little room for accountability for the misuse of these technologies. 

Similarly, the Ryder Review commission by The Ada Lovelace Institute found that the current legal framework was not ‘fit for purpose’ to regulate the use of the biometrics. 

The report also found a lack of independent oversight over the use of FRT and noted lack of human input in the use of the recognition technology. 

A Global Issue

The concern for Police’s use of AI facial recognition is global – the same technology used by the Metropolitan police was found to have misidentified Black men in the United States, leading to wrongful arrests. 

The EU and UN High Commissioners for Human rights have called for the prohibition of FRT in public spaces. 

In Scotland, the Parliament’s Justice Sub-Committee on Policing said that facial recognition technology was ‘not ‘fit for use’.

Overall Concerns 

Racial bias was a major concern flagged by the report: FRT disproportionately misidentified marginalised groups including women, people of colour and people with disabilities. FRT could therefore exacerbate the already existing bias in policing found in the UK. 

The UK Police has been under fire from human rights protesters for infringements on privacy and the freedom of expression. 

The report warned of the potential fall out of the increased use of FRT in police investigations, with the fear of potential consequences. 

After the passing of the new Police, Crime, and Sentencing Courts Act 2022, which gave police more power in restricting protestors, the use of live FRT could make the public more fearful of public political engagement. 

Recommended

Another concern raised by the report is the collaboration of Police with ethically dubious private technology firms. The Met Police’s Live Facial Recognition technology is sources from NEC Corporation, the same company used by US federal agents to target immigrant groups. 

“To protect human rights and improve accountability in how technology is used, we must ask what values we want to embed in technology,” said the report’s lead author Radiya-Dixit.

Due to the gaps in regulation and a failure to meet their ethical standards, the report led the researchers to join the growing call to ban the police’s use of facial recognition in public spaces.

Get the latest news from DIGIT direct to your inbox

Our newsletter covers the latest technology and IT news from Scotland and beyond, as well as in-depth features and exclusive interviews with leading figures and rising stars.

To subscribe, click here.