Last week, Asian Voice reported on the alarming rise of far-right sentiments and the increasing normalisation of racist attacks in daily news. Now, a new report sheds light on how predictive policing technologies are reinforcing racial discrimination within UK law enforcement.
Amnesty International UK’s report, ‘Automated Racism – How Police Data and Algorithms Code Discrimination into Policing’, reveals that predictive policing systems, used by at least 33 police forces—including the Met, West Midlands, and Greater Manchester—violate both national and international human rights obligations.
In the year ending March 2023 there were 24.5 stops and searches for every 1,000 Black people, 9.9 stops and searches for every 1,000 people with mixed ethnicity, 8.5 for every 1,000 Asian people – and 5.9 for every 1,000 white people.
The report presents multiple case studies highlighting how these technologies function across the UK. In London, for instance, the Metropolitan Police Service employs the Violence Harm Assessment Tool, which profiles individuals based on unverified intelligence reports, effectively labelling some as suspects without any criminal activity. In Lambeth Borough, between 2020 and 2021, the area recorded the second-highest stop-and-search volume in London.
Black individuals were four times more likely to be stopped than their white counterparts, with 80% of these searches leading to no further action. In Southwark Borough, Black people were 3.3 times more likely to be stopped than white people. The use of force by police was also disproportionately directed at Black individuals, with 45% of recorded incidents involving Black or Black British people.
When Asian Voice reached out to the Metropolitan Police regarding these statistics, a spokesperson stated, “The Violence Harm Assessment (VHA) is a Met Police intelligence tool to risk assess individuals that are suspected of being involved in violence across London. At present, all individuals on the VHA have been named as a suspect for at least three violent offences – they are not on the VHA because of intelligence alone. No stop and search data is used in the compiling of the VHA unless the stop and search resulted in the arrest of the person searched. Further information can be found on the Met website.”
The Mayor of London, Sadiq Khan, also provided a statement through his spokesperson, affirming his commitment to a fairer London stating, “The Mayor is committed to building a safer and fairer London for everyone and is working closely with the Met Police to address community concerns about disproportionality in the use of certain police powers.
“Sadiq is supporting the Met's Race Action Plan and ongoing work to build a Met that is anti-racist, inclusive, diverse and representative of the city it serves. His London Policing Board brings together a wide and diverse range of expertise and lived experiences to oversee and scrutinise the urgent reform of the Met.
“Technology and data-led intelligence are important policing tools to help the Met bear down on crime. The Mayor, MOPAC and the London Policing Board will continue to monitor their use to ensure it is both effective and fair.”
According to the report, West Midlands Police has also adopted predictive crime mapping under the Home Office’s ‘Grip’ program, designed to forecast knife crime and serious violence. However, the system proved to be inaccurate 83% of the time, misidentifying high-risk areas. In the West Midlands Police area, Asian people were 2.5 times more likely to be stopped and searched than white people in 2020.
Essex Police apparently utilises the Knife Crime and Violence Model, which the report criticises for criminalising individuals by association rather than evidence. The system factors in mental health and drug use data, disproportionately targeting individuals with health issues. Black individuals in Essex were nearly three times more likely to be stopped than white people, with disparities reaching six to seven times in some areas.
Greater Manchester Police’s XCalibre database has also come under scrutiny for disproportionately profiling Black individuals. The report highlights that GMP targeted Black cultural events, requiring venues to report attendees' ethnic backgrounds, effectively restricting freedom of association. People have been banned from events based on suspected gang affiliations without legal justification.
Sacha Deshmukh, CEO of Amnesty International UK, emphasised the human rights violations associated with predictive policing. “Policing should not serve to divide us, erode our rights, or contribute to racism. No matter our postcode or the colour of our skin, we all want our families and communities to live safely. The use of predictive policing tools violates human rights. There is no clear evidence that these technologies enhance safety, but there is overwhelming proof that they infringe on fundamental rights.”
He further warned of the systemic bias in these technologies. “These systems create a reality where technology determines criminality based purely on racial or socio-economic factors, leading to targeted policing and discrimination. Areas with high Black and racialised populations are disproportionately over-policed, generating biased police records. These records then feed into predictive policing tools, reinforcing a cycle of discrimination and criminalisation. This ‘feedback loop’ results in the over-policing of specific communities, perpetuating racial injustice.”
The report must be taken seriously by those in power, law enforcement, and communities alike. If the police themselves are part of the problem, how can we expect to find a solution in a multicultural society?
Moreover, when BAME communities are unfairly targeted, even by the police, it only fuels fear, reinforces harmful stereotypes, and leaves these communities more vulnerable to racist attacks. And let’s be honest, should the predictive policing system really be targeting ethnic communities when 81% of the prison population is white?
