3
0
In the modern day, Facial Recognition Technology (FRT) follows us everywhere. We might not know what it is, or how it works, but it is almost guaranteed we will encounter it in some form across our lives – be it day-to-day through unlocking our phones, or occasionally through airport customs.
But what about the technology we are less aware of?
The unthinkable happened to Robert Williams in 2020 when he was arrested for a crime he had not committed. Detroit officers arrested him in front of his children based on a tip that showed him as the closest match to the perpetrator they had caught on CCTV. At his questioning, he asked the officers, ‘I hope y’all don’t think all black people look alike,’ to which they said, ‘the computer says it’s you.’
This traumatic experience that Williams endured was due to the usage of AI FRT, which had access to driver’s licenses in Michigan – crucially, without the permission of the owners. Joy Buolamwini predicted these types of false arrests would occur back in 2016 – as a computer scientist, Joy looked at how FRT recognised different types of faces during her master’s degree. From this, she made the assertion that FRT was 32 times more likely to make a mistake identifying the face of a black woman than a white man.
Errors such as this occur due to the data the AI trains on being biased. This, with reference to the case of Robert Williams, is alarming considering that police in the United States are using it to catch criminals.
Perhaps more alarmingly, similar FRT is in use in the United Kingdom’s police force; Retrospective Facial Recognition (RFR) is confirmed to be in use by: South Wales Police, Gwent Police, Cheshire Constabulary, and the Metropolitan Police Service as of 2023. RFR allows police to look back in time to determine who was at a particular location, at a particular time, severely limiting individuals’ ability to live their day-to-day lives anonymously.
The case of R (Bridges) v Chief Constable of South Wales Police [2020] EWCA Civ 105 demonstrates how this use of FRT can be found to be in breach of our human rights. Here, the claimant brought judicial review proceedings against South Wales Police after they launched a surveillance project using FRT, which was deployed at certain locations where crime was likely to occur, capturing up to 50 faces per second of members of the public. It has been estimated that over the 50 deployments that were undertaken in 2017 and 2018, the technology may have captured sensitive facial biometric data from around 500,000 people, without their consent.
On the face of this, it is apparent that there are issues with the deployment of this technology; notably in respect of human rights violations. The Court of Appeal laid down in their judgement that the deployment of this technology was in breach of Article 8 of the ECHR (the right to respect for our private and family lives), and that the police were silent as to the risks to other rights which are likely to be affected by its use, namely the rights to freedom of assembly under Article 11 of the ECHR and freedom of expression under Article 10 of the ECHR.
This judgment is significant for several reasons; alongside the court naming the human rights violations that can come with deploying this technology, they also list the rights that could be violated, illustrating the court’s proactivity and anticipation of other issues. The effect of this pervasive surveillance has profound consequences to an individual’s human rights, most notably in respect to the development of their identity and their political participation.
This consequence is known as a chilling effect, which is where individuals modify their behaviour due to the fear of surveillance. This can influence a range of issues, such as what websites a person visits, what organisations they join, or whether they partake in political activities.All of which are worrying restrictions of freedom.
This usage of Facial Recognition Technology marks a change in police surveillance that may fundamentally alter the balance of power between the state and its citizens. The changing nature of AI is sure to be a central subject of legal development in the coming years – and where it will lead the law next is certain to be a captivating evolution.
Average Rating