Facial Recognition

WHAT IS FACIAL RECOGNITION TECHNOLOGY?

Since 2015, police forces in the UK have increased their use of facial recognition technology (FRT), a mass surveillance tool that can track and identify us as we go about our daily lives.

Despite Liberty client Ed Bridges winning the world’s first legal challenge to police use of the technology in 2020, every force in the country is now using some form of FRT.

How does facial recognition work?

Facial recognition works by taking live camera feeds, video and images and analysing a person’s facial features. It ‘maps’ distinct points of our faces and allocates a numbered code to that map, which is then compared to the codes of faces on police watchlists or databases.

There are three types of facial recognition:

1. Live Facial Recognition (LFR): This usually involves parking vans equipped with facial recognition cameras in places like high streets and shopping centres and scanning passersby in that area.
2. Retrospective Facial Recognition (RFR): This involves analysing existing videos or images – including CCTV, mobile footage, or doorbell cameras – and matching people to police databases.
3. Operator Initiated Facial Recognition (OIFR): Currently the least used of the three, this involves scanning people’s faces using a police officer’s mobile device to help discern their identity in real time.

Walking past a facial recognition camera is like showing your ID at a check point – something most of us wouldn’t want to have to do when we’re going about our daily lives.

How are the police using facial recognition technology?

Police forces were barely using this tech five years ago. Now, it’s widespread across the country.

It’s being deployed in our high streets; at sporting events like football and rugby; at Formula 1; at music festivals; at the seaside; and as a tool of intimidation at protests. It’s even being used in shops, with information passed to the police to track down people forced to shoplift during a cost-of-living crisis.

In the last year alone (July 2025 – July 2026), we’ve seen: 

• The Metropolitan Police double its number of weekly LFR deployments across London.
• A fleet of 10 new facial recognition vans that have allowed deployments to happen in places like Bedford and Manchester.
• South Wales Police setting up a network of facial recognition cameras for events like the Six Nations rugby, and trialling OIFR on officer’s phones.
• Hammersmith and Fulham approving plans for LFR cameras across ten locations in the borough, and for 500 CCTV cameras to be equipped with RFR capabilities.
• The British Transport Police conducting a six-month pilot of FRT cameras in train stations around the country.
• The UK’s first permanent facial recognition cameras going live in Croydon, South London.

That’s not all. In June 2026, the Metropolitan Police announced plans to place ‘static’ facial recognition cameras in London’s West End and Soho by the end of 2026, with further rollout across the capital in 2027. This – combined with the Met’s announcement to expand AI tools and drone usage across the capital – is a huge escalation in policing tools.

Why should we be concerned?

The law on facial recognition is being far outpaced by police use – which is expanding all the time.

Liberty Investigates, our independent journalism unit, has spent years exposing major scandals involving this technology.

In the past year alone, they’ve uncovered that: 

• At least six police forces are using facial recognition to track children as young as 12, with some not keeping records to explain why they are on those watchlists in the first place.
• A man named Alvi Choudhury was wrongly arrested for a burglary he never committed, in a town he’d never visited, because he was flagged by retroactive facial recognition (RFR).
• Essex Police paused using facial recognition cameras after a University of Cambridge study found racial bias within the technology.
• Some UK police forces lobbied to continue using an FRT system known to be biased against women, young people and members of ethnic minority groups, because the more accurate version produced fewer matches.
• Some police forces have been using ‘photofit’ images, similar to police sketches, in retrospective facial recognition scans.
• And others are expanding their use of ‘quick-check’ facial recognition apps, allowing officers to conduct OIFR searches on their phones.

There should be clear and consistent rules around how police are using facial recognition to ensure the rights of the public are protected. But currently, forces are being left to choose how they want to use this technology and essentially marking their own homework.

At the end of 2025, the Government finally announced a public consultation on police use of facial recognition technology and how it should be governed, which Liberty responded to. As it stands, we’re still waiting for an outcome to this consultation.

More recently, the King’s Speech in May 2026 announced plans for a dedicated legal framework to govern FRT; a welcome step after years of campaigning. But there has not yet been legislation laid before Parliament.

Any tech that has the potential to infringe on our rights, in the way scanning and identifying millions of people does, needs to have robust safeguards.

What are we calling for?

Facial recognition affects all of us. We should all be able to go about our daily lives without being tracked and monitored by the police.

Police forces have insisted that the public support the use of this technology, but recent polling suggests that more people than not oppose FRT being used as it currently is – to scan the faces of people walking in high streets and town centres, despite there being no imminent threats. 41% feel that it should only be used once laws are in place. (Deltapoll, July 2026)

The Government must halt the rapidly increasing roll out of facial recognition technology, follow the example of other countries and put laws, and robust safeguards in place to prioritise and protect the rights of the British public.

• live facial recognition is not used without independent sign-off
• police can only use live facial recognition to:
  • search for missing persons or victims of abduction, human trafficking and sexual exploitation
  • prevent an imminent threat to life or people’s safety
  • search for people suspected of committing a serious criminal offence
• facial recognition watchlists only contain images strictly relevant to the purposes above
• police give the public at least 14-days advance warning of live facial recognition deployments