Open consultation

Options assessment: Consultation on a new legal framework for law enforcement use of biometrics, facial recognition and similar technologies

Published 4 December 2025

Title: Consultation on a new legal framework for law enforcement use of biometrics, facial recognition and similar technologies

Type of measure: Legislation

Department or agency: Home Office

OA number: HO COA 1011

Contact for enquiries: fr-consultation@homeoffice.gov.uk

Date: 30 October 2025

1. Summary of proposal

1. The proposal is to create a durable legal framework for law enforcement’s use of biometrics, facial recognition and similar technologies.

2. Law enforcement organisations can use facial recognition and similar technologies to assist with:

• Supporting the location and arrest of people wanted for criminal offences.

• Preventing people who may cause harm from entering an area (for example, fixated threat individuals, persons subject to football banning orders).

• Supporting the location of people about whom there is intelligence to suggest that they may pose a risk of harm to themselves or others (for example, stalkers, terrorists, missing persons deemed at increased risk).

• After-the-event searching of images of suspects (for example, caught on CCTV, doorbells, phones) against a database of images post event/crime, to identify them.

• Checking the identity of an individual by searching their image against a police database of images.

3. Whilst these use cases are important to support the Safer Streets Mission^{[footnote 1]}, it is also important to ensure that any use of these technologies proportionately balances the potential benefits to public safety against the level of interference in individual rights such as rights to privacy and freedom of peaceful assembly (Human Rights Act,1998) and that there is sufficient oversight and governance of their use.

4. This consultation asks about principles that could be applied to police use of facial recognition, as well as a wider range of technologies, which all have the potential to interfere with people’s privacy and other rights such as freedom of expression and freedom of assembly and association. The Home Office will consider whether the new legal framework should extend to the use of other biometric and inferential technologies.

5. The legal framework should be easily understandable to the public and to the law enforcement bodies to whom it applies.

6. The proposals are for a new legal framework that will:

• be accessible and transparent – ensuring the police and the public can easily understand it and foresee how data and the technology will be used.

• set clear limits on police authority and discretion – defining who makes decisions and guarding against misuse.

• clearly define who sets rules and checks compliance – consolidating and simplifying oversight and regulation.

• be adaptive, allowing new technologies to be adopted consistently – making the framework future proof.

7. The Home Office is seeking consultation feedback on the following:

• The parameters of the framework, that is, the types of biometric and/or inferential technologies it should apply to.

• The types of law enforcement the framework should be applicable to.

• Assessment of interference in privacy and other rights caused by these technologies versus the seriousness of harm they seek to prevent.

• Necessary purposes for acquisition, retention, and use of biometrics and associated technologies

• Safe, fair, consistent uses of these technologies, including prevention of bias and discrimination.

• Circumstances in which access to images on government databases might be justified.

• Powers and remit of oversight and regulatory bodies.

8. The evidence from this consultation will support creation of a legal framework governing use of facial recognition and similar technologies for law enforcement. This will include clarifying oversight and regulation into a new body, which could also provide advice and guidance to law enforcement organisations.

2. Strategic case for proposed regulation

Legal clarity and oversight

9. The Court of Appeal in R (on the application of Bridges) v Chief Constable of South Wales Police (Information Commissioner and others intervening, 2020)^{[footnote 2]} ruled that there is a legal basis for the police to use facial recognition technology; this relies on common law powers and a patchwork of legislation, codes of practice and published police guidance^{[footnote 3]}. The patchwork of legislation means the rules are complex and difficult to understand, and may not give the police sufficient confidence to use the technology at greater scale. This complexity also limits public confidence that they will be used safely or fairly.

10. The former Minister of State for Policing, Dame Diana Johnson MP, held a series of roundtable meetings with the police, regulators, research institutions, civil society groups and industry. The consensus was that use of facial recognition and similar technologies would benefit from a clearer, specific legal framework, with simpler, clearer oversight. The former Minister of State for Policing also spoke in Parliament^{[footnote 4]} and held an information session with parliamentarians.

11. The oversight of law enforcement use of facial recognition and similar technologies is also a ‘patchwork’ of bodies and regulators –including: The Forensic Science Regulator (FSR), The Biometrics Commissioner and Surveillance Camera Commissioner, The Information Commissioners Office, Police and Crime Commissioners, Investigatory Powers Commissioner’s Office (IPCO), His Majesty’s Inspectorate of Constabulary and Fire Rescue Services (HMICFRS) and College of Policing. This patchwork makes oversight fragmented and difficult to understand.

Public confidence

12. The Home Office have commissioned a public attitudes survey to understand the public’s views about the use of facial recognition technologies in policing. The survey findings show that two in three people support police use of facial recognition technology. Respondents recognised benefits such as helping the police to catch criminals, enhance public safety, and locate missing or vulnerable individuals. However, concerns included the potential for misuse or hacking, risk of false identification, and issues around data privacy. This survey will be published alongside the consultation.

Law enforcement adoption

13. Technology is ever adapting and without government intervention there is a risk law enforcement fail to make the best use of all available technologies to aid in crime prevention and detection and alleviate resource pressures.

3. SMART objectives for intervention

14. The strategic objective will support the government’s Safer Streets Mission^{[footnote 5]} to reduce serious harm and increase public confidence in policing, whilst having positive impacts on the government‘s ambition to halve knife crime and halve violence against women and girls (VAWG) within a decade through the ability to identify suspects in a faster, more efficient manner.

15. Another strategic objective is to align with the government’s AI opportunities action plan^{[footnote 6]} whereby the government would like to ensure UK regulation is designed and delivered in a way that fuels fast, wide and safe development and adoption of AI. This will ensure that the framework is easily adaptable as technology grows, whilst attempting to minimise the risk of technology outgrowing law.

16. The Home Office consultation will result in a series of legislative measures to enable the safe, foreseeable, adaptable, consistent and proportionate use of facial recognition and similar technologies. This will be supported by guidance, investment in national systems and business change, and the delivery of national procurement frameworks.

17. More broadly, the objectives are for the new framework to:

• Objective 1 - be accessible and transparent for police and the public.

• Objective 2 - regulate the use of biometric and inferential technologies.

• Objective 3 - clearly define who sets rules and checks compliance – consolidating and simplifying oversight and regulation.

• Objective 4 - to ensure the police keep up with evolving technology, and make the most of that technology, whilst acting in accordance with the frameworks of guidance.

Measuring outcomes

18. The intended outcomes will increase confidence and trust in use of facial recognition in both the public, and law enforcement, aided by the consolidation of statutory oversight and intention to clarify regulation, in a single expert regulatory body. It will ensure that legal basis of use is well evidenced, sound and proportionate. The Home Office would also like to ensure use of technology to keep up with innovation whilst being within the parameters of the law.

4. Description of proposed intervention options and explanation of the logical change process whereby this achieves SMART objectives

Option 0: ‘Do nothing’

19. Maintain the existing legal framework and oversight bodies. Facial recognition and similar technologies will remain overseen by regulators (Information Commissioner’s Office and Equality and Human Rights Commission) and other oversight bodies (Biometrics and Surveillance Camera Commissioner, the FSR, HMICFRS, and the Independent Office for Police Conduct). The use of the technology will remain governed by the Data Protection Act 2018, Human Rights Act 1998, Equality Act 2010 and the Police and Criminal Evidence Act 1984, as amended by the Protection of Freedoms Act 2012.

Option 1

20. Create a new legislative framework for law enforcement use of facial recognition and similar technologies, including a single regulatory and oversight body. This option proposes the creation of a new legal framework focused on the use of facial recognition and similar technologies by law enforcement organisations. Under option 0, there would continue to be a legal basis for use of these technologies; legislation including the Data Protection Act 2018, Equality Act 2010, Police and Criminal Evidence Act 1984, Human Rights Act 1998, as well as the police’s common law powers, the College of Policing’s national guidance for live facial recognition, Surveillance Camera Code of Practice and the police’s own policies. However, the Home Office knows this does not give the police sufficient confidence to use it at significantly greater scale. Nor does it consistently give the public the confidence that it will be used responsibly according to clear rules. This is because the current framework is complicated and difficult to understand.

21. A new legislative framework would assist police in ensuring that they are using facial recognition and similar technologies in a safe consistent manner, within the parameters of the law.

22. Objective 1 will be achieved through designing and bringing forward a legal framework that allows for a more simple regime for the public and police to understand what the provisions are and how they will be applied.

23. Objective 2 will be achieved through publishing the legal framework and associated guidance. This will allow for clarity as to when the technology can be deployed, and who can deploy it.

24. Objective 3 will be achieved through clarifying oversight by creating a new regulatory and oversight body. This will ensure there is an easily identifiable single oversight body that the public and policing can refer to, for guidance, complaints or concerns on the use of such technologies. It will also allow for a body which can assess, and audit law enforcement use of these technologies in certain circumstances. This new body is likely to encompass and build upon the existing roles of the Biometrics and Surveillance Camera Commissioner and the Forensic Science Regulator in England and Wales.

25. Objective 4 will be achieved through making the legislative regime flexible both in terms of content and in terms of the powers needed to update it. For example, this could be outside of statute through codes of practice/guidance/the oversight structure. This would enable the police to safely and confidently adopt new technologies, aiding the police to prevent crime, catch criminals and keep Britain’s streets safe (objective 1).

5. Summary of long-list and alternatives

26. The consultation is taking place to inform the design of the options. Currently there are two options, to maintain the existing framework or to create a new framework. For Option 1 a number of aspects will be defined post-consultation, these are outlined in paragraphs 30 to 43. Options will be defined and short-listed based on the results of the consultation and further policy development.

27. Option 0: ‘Do nothing’ - Maintain the existing legal framework and oversight bodies – Facial recognition and similar technologies will remain overseen by current regulators, governed by the UK General Data Protection Regulation^{[footnote 7]} (UK GDPR) and the Data Protection Act 2018.

28. Option 1: Create a new legislative framework for law enforcement use of facial recognition and similar technologies, including a single regulatory and oversight body. This option proposes the creation of a new legal framework focused on the use of facial recognition and similar technologies by law enforcement organisations. The government proposes that:

• The framework would apply to law enforcement organisations including all police forces in England and Wales, and national and specialist law enforcement agencies like the British Transport Police and National Crime Agency and for law enforcement activity by other public bodies such as the Environment Agency, His Majesty’s Revenue and Customs (HMRC) or Border Force. The consultation will seek views on the extent to which the scope should be expanded to other organisations.

• Create a new regulatory and oversight body overseeing law enforcement use of biometrics, facial recognition and similar technologies, with statutory powers. This new body is likely to encompass existing regulatory bodies such as the BSCC and the FSR and additional functions as may be required in one organisation. This could include providing assurance that law enforcement use of biometric technologies is legal, responsible, and necessary. This new body could be responsible for:

  1. Regulating and overseeing law enforcement use of facial recognition and similar technologies.

  2. Issuing statutory codes of practice to which law enforcement have a duty to adhere.

  3. Overseeing quality and accuracy of the technology/algorithms to guard against bias.

• The exact size of the regulatory body is currently unknown and will be determined following the consultation. At a minimum it would be expected to encompass the FSR and BSCC, with additional powers mentioned above. As a maximum estimate, the IPCO would be a helpful comparison, given the potential powers of the body could extend to independently authorising and overseeing the use of powers, ensuring they are used in accordance with the law and in the public interest.

29. A number of aspects of the legal framework remain under consideration and will be defined based on consultation responses and further policy development. These include;

• Scope of application a) the types of technology, beyond facial recognition, which should be subject to the regime and b) the organisations to which a future framework should apply.

• Proportionate uses a) the relevant factors to consider when deciding when law enforcement should be permitted to use facial recognition, b) the levels of authorisation required and c) how a framework could protect against unjustified interference with an individual’s rights d) ensuring safe, fair, consistent uses of these technologies, including prevention of bias and discrimination.

• Searching public records a) when, if at all, it should be permissible for law enforcement organisations to conduct biometric searches of other government databases in circumstances where a person cannot be identified using police records and b) what appropriate safeguards would be.

Scope of application

30. Types of technology: The Home Office is seeking views on whether inferential technologies (for example, technology that looks for someone pacing a suicide hotspot) and object recognition matching technology should also be covered. The Home Office is considering extending the legislation to cover these technologies to reflect that new technology is constantly innovated and legislation should keep up. The Home Office are considering the extent to which the legislation can be flexible enough to be adaptable for future technologies.

31. Who it should apply to: The consultation proposes that the new legal framework will apply to law enforcement organisations. This includes all police forces in England and Wales, and national and specialist law enforcement agencies like the British Transport Police and National Crime Agency and for law enforcement activity by other public bodies such as the Environment Agency, HMRC or Border Force. The Home Office are seeking views on whether the new legal framework should only apply to law enforcement use or extend to other public and private uses of these technologies for example private companies identifying shoplifters, for example, Facewatch^{[footnote 8]}.

Proportionate uses

32. Balancing privacy concerns versus the seriousness of harm they seek to prevent: The circumstances in which such technologies can be used will be detailed in the new legal framework. This is to ensure clarity when these types of technologies can be used, including: voluntary or involuntary usage; usage in public or private spaces; overt or covert usage; real time or retrospective usage; access to the data and other types of assessments. However, whilst the Home Office is certain the answers to these assessments around interference should be included in the framework, there is uncertainty on the level of detail that should be included. The Home Office would like to consult on the aforementioned details to increase both policing and public trust and accountability.

33. Necessary purposes for acquisition, retention, and use of biometrics and associated technologies: The government is uncertain whether the Home Office should mandate if ‘seriousness’ should be a deciding factor on whether facial recognition and similar technologies should be used and set out in the new framework. The consultation asks how to define ‘seriousness’ and what factors are relevant to defining ‘seriousness’ to determine if it should be included in the framework.

34. A new legal framework would also need to prescribe when these technologies should be used, and when higher levels of authorisation should be needed.

35. This is to ensure that it is clear when these types of technology can or should be used, or when more oversight is needed to make deployment decisions. The Home Office would like to consult on these details to increase both policing and public trust and accountability.

36. Safe, fair, consistent uses of these technologies, including prevention of bias and discrimination: if a new legal framework for facial recognition and similar technologies is implemented, it is likely that the single oversight body should set rules to mitigate potential bias and discrimination that could arise from the usage of these technologies.

37. The Home Office would like to consult on whether the oversight body should set specific rules for law enforcement to follow to guard against bias and discrimination.

Circumstances in which access to images on government databases might be justified

38. If and when it should be permitted: The Home Office is considering whether this should be included in the framework and is consulting on whether searches should be permitted and, if permitted, in what scenarios, for example, for safeguarding, identification of deceased individuals.

39. Safeguards: It is likely the circumstances in which this could be done would need further scrutiny. The government are seeking views on what requirements or safeguards would be needed to be in place to enable these searches. The government are considering, for example, who should approve or review usage.

40. Powers and remit of oversight and regulatory bodies: as mentioned above if a new legal framework for facial recognition and similar technologies is implemented, the creation of a new single oversight body will form a part of that framework. This is to ensure that it is clear who has oversight over these types of technologies, or regulatory responsibility, increasing both public trust and accountability. The Home Office would like to consult on whether forensic regulation, biometric regulation and oversight of biometrics and surveillance are similar enough to be overseen by a single independent body.

41. The Home Office would also like to consult on what powers and or obligations this body should have to aid them in overseeing police use of facial recognition and similar technologies. This could include codes of practice detailing legal and ethical obligations when using the technology; investigating instances where technology is hacked or misused; receiving complaints from anyone; publishing an annual report detailing compliance with the relevant code(s) of practice.

6. Description of shortlisted policy options carried forward

42. Option 0: ‘Do nothing’ - Maintain the existing legal framework and oversight bodies – Facial recognition and similar technologies will remain overseen by current regulators, governed by the UK General Data Protection Regulation^{[footnote 9]} (UK GDPR) and the Data Protection Act 2018.

43. Option 1: Create a new legislative framework for law enforcement use of facial recognition and similar technologies, including a single regulatory and oversight body. As outlined above.

44. The Impact Assessment, which will be developed after the consultation will include details on the design of the new legal framework.

7. Monitoring and evaluation

45. If Option 1, ‘introduce a new legislative framework’ is taken forward, the Home Office intends to introduce legislation in accordance with the views given in response to the consultation. After implementation, the new regime will need to be in place for a minimum of three years to ensure compliance with the new regime rectifying any issues. After this time frame, the Home Office will consider whether to conduct a post-implementation review of the new legislation.

46. The Home Office has conducted a public attitudes survey – detailed above in paragraph 12 – which can be used to inform the post-implementation review, by widening the scope of the survey to include the legislative framework.

47. Success will be measured by the reaction of public, police and other stakeholders to the new legislation. Currently, the Home Office receives high volumes of correspondence from the public, Civil Society Groups, and MPs asking for changes to the current patchwork governing facial recognition, and the Home Office has faced Judicial Reviews due to discontent with the manner in which biometric and other inferential technologies are used. The Home Office will remain in close communication with the public, the police and other stakeholders to gauge their response and monitor their concerns or whether they are requesting further change to understand the success of the intervention.

48. The Home Office will assess whether the original objectives detailed above around regulation and compliance are met through monitoring data from police forces detailing the use of such technologies.

49. The Home Office will assess whether there have been any unintended consequences, including disproportionate or unexpected administrative costs through including a process evaluation in the post-implementation review.

8. Minimising administrative and compliance costs

50. Depending on the outcomes of the consultation, and the form of the proposed legislative framework, both administrative and compliance costs are likely to be incurred to get systems up to speed, or training personnel to comply with the new framework. These costs will largely fall to the Home Office and law enforcement agencies.

51. However, the Home Office has been taking steps already through business change activity to minimise any costs with adopting a new legislative framework for facial recognition and similar technologies. This can be seen through system upgrades for databases that hold biometrically identifiable information or recommending training in using facial recognition and similar technologies.

52. Declaration

Department: Home Office

Contact details for enquiries: fr-consultation@homeoffice.gov.uk

Minister responsible: Sarah Jones MP, Minister of State for Policing and Crime

I have read the Consultation Options Assessment and I am satisfied that, given the available evidence, it represents a reasonable view of the likely costs, benefits and impact of the leading options.

Signed: Sarah Jones MP (Minister for Policing and Crime Prevention)

Date: 06 November 2025

Summary: Analysis and evidence

Price base year: 2025/26

PV base year: 2025/26

	Option 0: ‘Do Nothing’	Option 1: Create a new legislative framework for law enforcement use of facial recognition and similar technologies.
Net present social value (with brief description, including ranges, of individual costs and benefits)	£0	Central Estimate -£0.5 million (-£0.2 million to -£0.8 million). Until the scope of the legal framework is set out, only the familiarisation costs and estimated running costs of the programme management team are considered in the NPSV.
Public sector financial costs (with brief description, including ranges)	£0	Familiarisation Costs of £39,800 (£11,300 to £83,400) and Programme Team Costs of £0.5 million (£0.2 million to £0.8 million).
Significant un-quantified benefits and costs (description, with scale where possible)		Non-monetised costs include implementation costs beyond the programme team, costs to the CJS and costs of the oversight body. Non-monetised benefits include increased productivity, increased positive crime outcomes and increased public trust and confidence as a result of the new legal framework.
Key risks (and risk costs, and optimism bias, where relevant)		The scope and size of the new legal framework are to be determined after the consultation, so estimates of monetised costs and benefits are uncertain.
Results of sensitivity analysis	Not applicable	Not applicable

Annex

A. Evidence Base

General Assumptions and Data

1. The general assumptions used in this COA which is in line with the guidance set out in HM Treasury (2022) Green Book:

• The appraisal period used is 10 years from 2025/26 to 2034/35

• The Price Base year used is FY 2025/26

• The Present Value base year 2025/26

• A social discount rate of 3.5 per cent is used to obtain present values

• All figures have been rounded to two decimal places

2. Data and assumptions have been drawn from several sources including:

• Internal Home Office pay data

• Reading Speed Data

Monetised impacts

3. The legal framework has not been drafted at the time of this appraisal. The government is consulting the public on scope of application of the legal framework, defining what would be considered proportional uses, and the circumstances under which access to images held in government databases may be justified.

4. At this stage it is not known how the framework will change the use of facial recognition technology in policing and what the impacts will be. It is not currently possible to quantify or monetise the costs of implementing the legal framework or to assess the potential impacts on the criminal justice system.

5. The monetised impacts include familiarisation costs and running costs of the new oversight body.

Costs

Familiarisation costs

6. Familiarisation costs refer to the reading and comprehension of any guidance documents needed for the legal framework to be implemented. The exact guidance needed will depend on the legal framework itself but it is expected that the guidance will be approximately 150 pages long. It is also expected that this guidance will be read by one superintendent and two inspectors per police force.

7. To estimate familiarisation costs, the estimated reading time of the information documents are multiplied by the labour costs of those required to read them. Using reading time estimates provided by ReadingSoft^{[footnote 10]} estimates, there is a range of reading time estimates of between 86 minutes and 631 minutes, with a central estimate of 301 minutes.

8. This estimated reading time was multiplied by the average hourly labour cost of a superintendent and two inspectors to estimate the total familiarisation cost per force. This was then multiplied by 43 for the 43 force areas.

9. Table 1 shows the estimated reading times in hours of the guidance, and the expected familiarisation costs of reading the guidance.

Table 1: Estimated breakdown of police familiarisation costs (2025/26 prices, £)

Superintendents

Estimate	Number of staff	Hourly rate	Reading speed (wpm)	Time spent reading (hours)	Cost
Low	43	75.12	700	10.52	4,600
Central	43	75.12	300	5.02	16,200
High	43	75.12	200	1.43	34,000

Inspectors

Estimate	Number of staff	Hourly rate	Reading speed (wpm)	Time spent reading (hours)	Cost
Low	86	54.67	700	10.52	6,700
Central	86	54.67	300	5.02	23,600
High	86	54.67	200	1.43	49,500

Source: Home Office Internal Analysis and Readingsoft.com

10. The total familiarisation costs can be seen in Table 2 below.

Table 2: Total estimated police familiarisation costs (2025/26 prices, £)

Scenario	Total cost
Low	11,300
Central	39,800
High	83,400

Source: Home Office Internal Analysis and Readingsoft.com

11. The total estimated familiarisation costs range between £11,300 and £83,400, with a central estimate of £39,800. Familiarisation costs are incurred in year one only.

Oversight body programme team costs

12. The government proposes combining the roles of the BSCC and the FSR in England and Wales and redefining the role to include regulation and oversight of facial recognition and similar technologies.

13. The size and structure of the organisation will be defined by the scope of the legal framework itself. The costs of the oversight body and programme team responsible for setting it up are uncertain at this stage.

14. The central estimate for the programme team costs is based on costs of programme teams for similar projects, and best internal assumptions. Due to the uncertainty around the size and structure of the organisation, the low and high estimates model these costs with varying team sizes and staffing levels.

15. Table 3 shows the estimated number of staff and their full labour costs (including pension and NI employer contributions) for the programme team. This assumes a three-person (low), six-person (central) and nine-person (high) team in each respective scenario.

Table 3: Estimated breakdown of programme team costs (£ million, 2025/26 prices)

Scenario	Grade	Number of staff	Annual cost
Low	HEO	2	0.11
Low	G7	1	0.10
Low	G6	0	0.00
Central	HEO	3	0.17
Central	G7	2	0.20
Central	G6	1	0.12
High	HEO	4	0.22
High	G7	3	0.30
High	G6	2	0.24

Source: Home Office Internal Analysis

16. The total estimated cost of the programme team for the initial set-up of the oversight body can be seen in Table 4 below.

Table 4: Total estimated programme team costs (£ million, 2025/26 prices)

Scenario	Total cost
Low	0.21
Central	0.49
High	0.76

Source: Home Office Internal Analysis

17. The running costs of the oversight body will change once the legal framework is established and the number of staff who will be needed is known. Costs will only be considered additional if they exceed the current running cost of the organisations the new body will encompass.

Benefits

18. There are no monetised benefits estimated.

Overall

Value for Money

19. The NPSV is estimated to be between -£0.2 million and -£0.8 million with a central estimate of -£0.5 million.

20. The negative NPSV is due to there being no monetised benefits at this stage and the monetised costs being familiarisation and programme team. There is no business impact, meaning no BNPV or EANDCB.

21. All costs are highly uncertain as they will depend on the legal framework’s scale and scope which will be designed based on consultation responses.

Non-monetised impacts

22. There are several non-monetised impacts of the introduction of a new legal framework.

Non-monetised costs

Costs to policing of implementation of the Legal Framework:

23. The costs of implementation for the legal framework will include the costs to policing resulting from new processes or changes in operations as a result of the new legal framework. Currently, given the legal framework has not been set out it is not possible to assess what these costs will be, so these costs are not monetised.

24. It is expected that the broader the scope and scale of the framework, the greater these costs will be. The Home Office will further refine these costs and attempt to monetise them following the consultation.

Costs to the Criminal Justice system

25. During engagement with policing stakeholders, it has been highlighted that legal concerns are limiting police use of use of facial recognition and similar technologies. It is expected that the legal framework will increase the police’s confidence in when this technology can be used and lead to increased use of it by the police.

26. Between January 2025 and September 2025, the Metropolitan Police reported over 700 arrests^{[footnote 11]} for offences including rape, domestic abuse, knife crime, grievous bodily harm, robbery, and 50 registered sex offenders in breach of their conditions^{[footnote 12]} as a direct result of using live facial recognition to locate people. In response to the summer disorder last year, the police report that the use of retrospective facial recognition led to 127 arrests.^{[footnote 13]} Home Office expects that the increased use of the framework will lead to more crime outcomes.

27. It is likely that there will be costs to the criminal justice system, including: cost of court cases, legal aid and any resulting additional sentences. This cannot be assessed without understanding the scope of the legal framework, and gathering further evidence on the effectiveness of this technology in aiding crime outcomes. Following the consultation, this will be considered further.

28. As with the costs of implementation it is expected that the broader the scope and scale of the legal framework as well as the crime types that are targeted through the use of facial recognition.

Costs of the oversight body

29. The government proposes creating a new oversight body to consolidate and clarify existing regulatory roles, ensuring responsible use of these technologies. This new body is likely to encompass the functions of the existing regulatory bodies such as the BSCC and the FSR, and additional functions as may be required in one organisation.

30. The costs of the oversight body beyond the programme team are not monetised, and the size, scope and structure of the body will be design based on the results of the consultation and in line with the legal framework.

Upper bound

31. The IPCO was created to increase public understanding of the oversight and accountability of investigatory powers. In the 2023 Post Implementation Review,^{[footnote 14]} the transition costs were estimated to be £3.1 million in 2022/23 prices, and the run costs were estimated to be £6.2 million on average for the first three years.

32. In 2025/26 prices the transition costs would be £3.4 million and the average running costs would be £7.0 million. This likely represents a rough order of magnitude upper bound for the costs of the new oversight body.

Lower bound:

33. The annual running costs of the BSCC are £0.65 million (2025/26 prices). The annual running costs of the Forensic Science Commissioner are £1.55 million (2025/26 prices).

34. The oversight body as a minimum is expected to include the existing responsibilities of both the BSCC and the FSR alongside additional responsibilities. The sum of the costs of both bodies, £2.2 million, represents a rough order of magnitude lower bound for the running costs of the new oversight body. Assuming transition costs would be proportionally the same as the IPCO in comparison to running costs, then transition costs would be £1.1 million.

35. Whilst these costs may indicate the potential scale of costs of the new oversight body, it is not known where in this range the costs will sit as it will depend on the design of the oversight body which has not yet been set. These costs will be further considered following the consultation, but costs will only be considered additional and included in this appraisal if they exceed the current running cost of the organisations the new body will encompass.

Non-monetised benefits

Productivity improvements

36. It is expected that this legal framework will give police greater confidence and lead to increased use of facial recognition and similar technologies (See paragraph 25). There is evidence to suggest that increased use of facial recognition and similar technologies will lead to increased productivity, as identifications can be done in minutes, rather than weeks^{[footnote 15]}.

37. The Metropolitan Police have estimated that personnel costs per arrest using live facial recognition are up to 25 per cent lower than traditional tactics and South Wales Police estimate that retrospective facial recognition (RFR) technology cuts identification time from an average of fourteen days to minutes. Further, the accuracy benefit of these technologies can be seen in police data on live facial recognition technology use in 2025, where 0.6 per cent of alerts made by the technology resulted in a false identification.^{[footnote 16]}

38. The expected productivity improvements have not been quantified or monetised as the use of these technologies will depend on how the legal framework is designed. This will be defined and monetised based on the responses to the consultation.

Increased positive crime outcomes

39. It is expected that this legal framework will lead to increased use of facial recognition and similar technologies (See paragraph 25). Increased use of these technologies is expected to lead to more positive crime outcomes through apprehending people wanted for serious crimes, who might not have been caught otherwise. Increased numbers of positive crime outcomes is expected to reduce crime.

40. For example, data on police use of live facial recognition technology in 2025 shows that, on average, 46 per cent of live recognition alerts led to arrests.

41. It is not possible to monetise the benefit of increased positive crime outcomes as it is not known what the legal framework will be, and how much the use of these technologies will increase. Furthermore, the scale and scope of the legal framework will determine what crime types are targeted which will impact the corresponding crime benefit.

Increased public trust and confidence

42. Several previous surveys have indicated high levels of public support for the use of facial recognition technology in policing, with around 70 per cent of the UK population being supportive of its use by the police in criminal investigations^{[footnote 17]}. Despite this, the same survey found that 46 per cent of people think the public should be able to opt out of or consent to facial recognition technology.

43. The new legal framework could further improve public attitudes towards the fair use of facial recognition and similar technologies. This change in public confidence has not been monetised however, as the specifics of the legal framework are not yet known, so the extent to which individual rights and privacy will be protected is uncertain. Additionally, the scope of the framework will determine the extent to which the technologies can be used, which will also impact public confidence and trust.

Expected overall impacts

44. The overall impacts are expected to be greater confidence in the use of facial recognition and other similar technologies from both the public and the police. It is likely this will lead to increased use of this technology which is expected to result in productivity gains and increased positive crime outcomes. There will likely be some costs of implementing the framework and the regulatory body, but these are not currently possible to monetise as they will depend upon the design of the legal framework.

Distributional impacts

45. This legal framework is expected to ensure consistency across police forces. There are not considered to be any significant distributional impacts from the proposed legislation.

Impacts on wider government priorities

Business Environment

46. There are not considered to be any significant wider business environment considerations from the proposed legislation.

International Considerations

47. There are not considered to be any international considerations from the proposed legislation.

Natural capital and Decarbonisation

48. The proposed legislation will not impact the UK’s natural carbon or decarbonisation.

B. Statutory Equalities Duty

All Consultation OAs are required to have the Statutory Equalities Duty reviewed by the SRO before signoff.

Mandatory specific impact test - Statutory Equalities Duties: Complete

The proposals on police use of facial recognition and similar technologies involve the creation of a new legislative framework and single oversight body to govern use. The Home Office gave due regard to the Public-Sector Equality Duty when considering all options and developing the proposal.

The Home Office intends to hold a wide public consultation to seek views from the public, and all other stakeholders on both options under consideration to understand both public and policing views on the proposed amendments and to learn of the impact of the proposals and of any potential unintended consequences. The Home Office has decided that a public consultation, rather than a targeted consultation, would be the most effective way to gather the best evidence to review the current framework and proposed changes to ensure changes are in the best interest of all effected.

There is currently known bias in the current use of facial recognition and similar technologies (sex, race, religion and age), however the purpose of the proposals set out above have been carefully considered to mitigate these known biases. The proposals around legislative changes and better, more consolidated oversight are designed to ensure consistency in facial recognition use, increased independent algorithm testing and oversight governing all use of AI enabled technologies. The Home Office does consider that any impacts on these characteristics are justified by the overall benefits provided by facial recognition and similar technologies to policing and public safety. However, the proposed measures will ensure better oversight and governance of technologies, taking into account proportionality considerations.

The SRO has agreed these summary findings.

1. Safer Streets Mission - GOV.UK: https://www.gov.uk/missions/safer-streets ↩

2. R (on the application of Bridges) v Chief Constable of South Wales Police (Information Commissioner and others intervening) case judgment https://www.judiciary.uk/wp-content/uploads/2020/08/R-Bridges-v-CC-South-Wales-ors-Judgment.pdf ↩

3. UK General Data Protection Regulation (UK GDPR) the Data Protection Act 2018, the Human Rights Act 1998, the Equality Act 2010, the Surveillance Camera Code of Practice ↩

4. Westminster Hall Debate – Hansard: https://hansard.parliament.uk/Westminster%20Hall%20Debate/2024-11-13/debates/E334DF95-2313-4AAC-AA25-D34F8F7C8DD5/web/ ↩

5. Safer Streets Mission - GOV.UK: https://www.gov.uk/missions/safer-streets ↩

6. AI Opportunities Action Plan - GOV.UK: https://www.gov.uk/government/publications/ai-opportunities-action-plan/ai-opportunities-action-plan ↩

7. UK General Data Protection Act – GOV.UK: https://www.legislation.gov.uk/eur/2016/679/contents ↩

8. Facial recognition for retail security: https://www.facewatch.co.uk/facial-recognition-for-retail-sector/ ↩

9. UK General Data Protection Act – GOV.UK: https://www.legislation.gov.uk/eur/2016/679/contents ↩

10. ReadingSoft reading speed calculator – readingsoft.com ↩

11. Live facial recognition deployments and outcomes data - https://www.met.police.uk/SysSiteAssets/media/downloads/force-content/met/advice/lfr/deployment-records/live-facial-recognition—deployment-record-2025-to-date2.pdf ↩

12. BBC News: More than 700 arrests from live facial recognition (2025) - https://www.bbc.co.uk/news/articles/c62lq580696o ↩

13. National Police Chiefs Council: NPCC response to Inspectorate report on summer disorder (2024) - https://news.npcc.police.uk/releases/npcc-response-to-inspectorate-report-on-summer-disorder ↩

14. Investigatory Powers Act (2016) Post Implementation Review - [https://assets.publishing.service.gov.uk/media/6894a00ae7be62b4f064310f/Annex_A_-2023-04-28_IPA_Review_PIR.pdf](https://assets.publishing.service.gov.uk/media/6894a00ae7be62b4f064310f/Annex_A-_2023-04-28_IPA_Review_PIR.pdf) ↩

15. Police use of Facial Recognition: Factsheet - https://homeofficemedia.blog.gov.uk/2023/10/29/police-use-of-facial-recognition-factsheet/ ↩

16. Metropolitan Police live facial recognition deployments 2025 data - https://www.met.police.uk/SysSiteAssets/media/downloads/force-content/met/advice/lfr/deployment-records/live-facial-recognition—deployment-record-2025-to-date2.pdf ↩

17. Beyond face value: public attitudes to facial recognition technology - https://www.adalovelaceinstitute.org/wp-content/uploads/2019/09/Public-attitudes-to-facial-recognition-technology_v.FINAL_.pdf ↩