🔗 Share this article

Law enforcement agencies across the UK successfully lobbied to deploy a facial recognition system known to be discriminatory against females, youths, and individuals from ethnic minority groups, after complaining that a less biased version generated a reduced number of investigative leads.

The Technology in Practice

British police utilize the national police database to carry out searches using historical face recognition. This process involves comparing a “probe image” of a person of interest against a repository of over 19 million mugshots to identify potential matches.

Admitted Bias

The UK interior ministry conceded last week that the technology was flawed. This acknowledgment followed a review by the government's National Physical Laboratory determined it incorrectly matched people of Black and Asian heritage and women at much greater frequency than Caucasian males. The Home Office said it “had acted on the findings”.

“This raises the question of whether this technology only becomes effective if users accept biases in ethnicity and sex. Operational ease is a poor argument for disregarding basic freedoms.”

Long-Standing Problem

Internal documents reveal that this discriminatory flaw has been recognized for over twelve months. Furthermore, law enforcement argued to overturn an earlier ruling that was designed to address the problem.

Police bosses were notified of the algorithmic discrimination in September 2024. The Home Office-commissioned NPL review found the system was more likely to produce incorrect matches for images depicting women, Black people, and those under 40 years old.

A Reversed Decision

In response, the national police leadership body mandated that the confidence threshold required for potential matches be raised to a point where the disparity was significantly reduced.

However, this directive was overturned the following month following complaints from police that the modified technology was producing fewer “useful lines of inquiry”. Internal records indicate the stricter setting reduced the number of searches that yielded possible identifications from over half to a just under 15%.

Profound Inequalities

Although the Home Office and NPCC refused to say what threshold is now in operation, the latest NPL study found the system could generate incorrect matches for women of Black heritage almost 100 times more frequently than for white women at specific configurations.

The Home Office commented on these findings: “The testing found that in a specific scenarios the software is has a greater tendency to incorrectly include some population segments in its search results.”

Balancing Utility and Fairness

Describing the effect of the brief increase to the system's confidence threshold, the police records note: “The change greatly lessens the impact of discrimination across protected characteristics of race, age and sex but had a significant negative impact on operational effectiveness”. The documents further note that forces complained that “a once effective tactic returned results of questionable value”.

Broader Rollout Plans

Meanwhile, the UK administration has opened a two-and-a-half-month consultation on its plans to widen the use of facial recognition technology. The minister for police the relevant minister has described the tool as the “most significant advance since DNA matching”.

Expert and Oversight Concerns

The chair of a police oversight board, head of the advisory panel for the national policing equality strategy, said: “There was very little discussion in race action plan meetings of the facial recognition rollout even with clear relevance with the plan’s concerns.

“These revelations show once again that the pledges to combat discrimination the police has undertaken via the race action plan are failing to be integrated into broader operations. Independent assessments have warned that innovative tools are being rolled out in a context where racial disparities, inadequate oversight and poor data collection already persist.

“All deployment of facial recognition must adhere to rigorous official guidelines, be subject to external review, and demonstrate it diminishes rather than exacerbates ethnic bias.”

Home Office Response

A government representative stated: “The Home Office treat the findings of the study with utmost gravity and we have implemented changes. A new algorithm has been externally evaluated and acquired, which has no statistically significant bias. It will be tested early next year and will be undergo evaluation.

“Our priority is ensuring public safety. This gamechanging technology will assist police to put criminals and rapists behind bars. There is human involvement in every step of the procedure and no arrest or charge would be pursued without trained officers carefully reviewing the output.”