Home / News / Criminal Defence / Defending Against 2026 Facial Recognition Evidence in Public Order Trials 

Defending Against 2026 Facial Recognition Evidence in Public Order Trials 

09 Jun, 2026 | TSABI Team
Table of Contents
Do you have more questions?

Do you have more questions?

Cant find the answer youre looking for? Please chat to our friendly team.

Get In Touch

Table Of Contents

The Legal Status of Facial Recognition Evidence in 2026  +
Live Facial Recognition (LFR) vs. Retrospective Identification 
The Threshold of "Similarity Scores" 
Grounds for Challenging Biometric Identification  +
Identifying Demographic Bias and High Error Rates 
Scrutinising Watchlist Integrity and Proportionality 
Procedural Defences and Human Rights Arguments  +
Failure to Provide Public Notice 
Challenging the "Necessity" of Surveillance 
Why Expert Legal Advocacy is Essential for Surveillance Cases  +
How TSABI's Defence Solicitors Can Assist You 

In 2026, the use of Live Facial Recognition (LFR) and Retrospective Facial Recognition (RFR) by police forces across England and Wales has reached unprecedented levels. While the Home Office and regional forces promote this technology as a vital tool for maintaining public order, the evidence it produces is far from infallible. At TSABI, we stand at the forefront of legal challenges against biometric surveillance, ensuring that AI-driven evidence is scrutinised with the same rigour as any other form of testimony. 

Our visibility as a leading criminal defence firm allows us to monitor the rapid deployment of LFR vans and permanent camera networks. We provide inclusive legal support for all individuals, recognising that mass surveillance often impacts diverse communities disproportionately. Our mission is to ensure that a computer-generated “match” is never the final word in a court of law. 

Facial recognition technology is a biometric tool that converts an image of a human face into a numerical template to compare it against a “watchlist” of suspects. At TSABI, our experts provide this service by deconstructing the specific technology used in your case to identify flaws in the prosecution’s identification. 

In 2026, despite extensive government consultations, there is still no single “Facial Recognition Act.” Instead, police rely on a patchwork of powers, including the Data Protection Act 2018 and the College of Policing’s Authorised Professional Practice (APP). We argue that this lack of specific legislation makes the current deployment of such technology open to significant legal challenge. 

Live Facial Recognition (LFR) vs. Retrospective Identification 

It is vital to distinguish between how the authorities gather biometric evidence, as the defence strategy changes for each: 

  • Live Facial Recognition (LFR): Cameras scan passers-by in real-time at events like protests or football matches. If the software flags a match, an officer makes an immediate intervention. 
  • Retrospective Facial Recognition (RFR): Police run footage from CCTV, doorbells, or social media through an algorithm after an incident to identify participants. 

We scrutinise the source of the footage and the digital chain of custody for both methods to ensure your rights remain protected. 

The Threshold of “Similarity Scores” 

Facial recognition software does not provide a simple “yes” or “no” answer. Instead, it generates a similarity score (usually between 0 and 1). In many 2026 systems, police forces set a threshold such as 0.6 to trigger an alert. We argue that these thresholds are often set too low to prevent “false positives,” especially in crowded or poorly lit public order scenarios where the software is most likely to make mistakes. 

Grounds for Challenging Biometric Identification 

Challenging facial recognition evidence requires a technical and a legal approach. We focus on the “black box” nature of the algorithms used by law enforcement to prove they are not as reliable as the police claim. 

Identifying Demographic Bias and High Error Rates 

In 2026, independent research continues to highlight that certain algorithms perform less accurately on Black and Asian faces, as well as on women and minors. At TSABI, we use expert witnesses to demonstrate to the court that the technology may possess a built-in bias. This bias can lead to a wrongful identification, and we believe no one should face conviction based on flawed, discriminatory software. 

Scrutinising Watchlist Integrity and Proportionality 

For an identification to be lawful, the “watchlist” used by the police must be specific and necessary. We challenge cases where the watchlist was excessively broad or included individuals who should have been removed long ago. We argue that the entire deployment was an unlawful interference with Article 8 (Right to Privacy) of the European Convention on Human Rights. 

Environmental Factors: Lighting, Angles, and Obstructions 

Real-world conditions rarely match laboratory testing. We analyse the original footage for: 

  • Low resolution or “grainy” images that lose facial detail. 
  • Poor lighting, deep shadows, or extreme weather conditions. 
  • Partial obstructions, such as face coverings, glasses, or scarves. 

If the input image was poor, the resulting match is legally unreliable and we move to have it excluded. 

Procedural Defences and Human Rights Arguments 

Beyond technical flaws, we examine whether the police followed the correct procedures for surveillance. If the authorities cut corners, the evidence they gathered may be inadmissible. 

Failure to Provide Public Notice 

The law requires the police to make the public aware when LFR is in use. If the police failed to display clear signage or provide social media notification in 2026, the evidence gathered may be ruled inadmissible under Section 78 of PACE as a breach of procedural fairness. We investigate the deployment site to ensure the police met their transparency obligations. 

Challenging the “Necessity” of Surveillance 

Under the Human Rights Act 1998, the use of facial recognition must be proportionate. We argue that using mass biometric scanning for minor public order offences is an excessive use of state power that outweighs the public interest. Our team fights to ensure that surveillance remains a last resort, not a default policing tool. 

Facial recognition evidence often carries an aura of “scientific certainty” that can unfairly influence a jury. Without a specialist defence, this evidence can lead to wrongful convictions. 

How TSABI’s Defence Solicitors Can Assist You 

At TSABI, our expert solicitors and barristers offer a unified defence strategy. We provide: 

  • Digital Forensic Analysis: We work with technical experts to audit similarity scores and algorithm performance. 
  • Direct Advocacy: Our barristers are experienced in cross-examining police “Face ID” technicians to expose the limitations of their tech. 
  • Proactive Defence: We seek to exclude biometric evidence at the pre-trial stage to protect you from a long trial. 
  • Inclusive Support: We ensure that every client receives a robust defence against the biases inherent in modern surveillance technology. 

If you have been identified via facial recognition or are facing a public order trial, contact TSABI today for professional and authoritative legal representation. 

Get Started Today

Cant find the answer youre looking for? Please chat to our friendly team.

Search

Find articles, FAQs, lawyers, and more

Press Enter to search or ESC to close

Secret Link