{ "metadata": { "schema_version": "1.0.0", "generated_at": "2026-06-29T16:29:54+00:00", "baseurl": "/ai-mapping", "total_tools": 70, "total_organisations": 48, "criminal_justice_stages": { 1: "Community Policing and Offender Management", 2: "Intelligence", 3: "Investigation", 4: "Charging Decision or Alternative Disposal", 5: "Trial or Guilty Plea", 6: "Sentencing", 7: "Prison and Parole", 8: "Probation" }, "deployment_stages": ["Trialled","Live","Experimental","Stage unknown"], "development_types": ["In-house","Third-party","Academic collaboration","Unknown"], "taxonomy_inference_modes": ["Analysis","Synthesis","Generation"] }, "organisations": [ { "id": "avon-and-somerset", "name": "Avon and Somerset", "url": "/ai-mapping/organisations/avon-and-somerset/", "kml_url": "/ai-mapping/assets/kml/avon-and-somerset.kml", "tools_count": 5 }, { "id": "bedfordshire", "name": "Bedfordshire", "url": "/ai-mapping/organisations/bedfordshire/", "kml_url": "/ai-mapping/assets/kml/bedfordshire.kml", "tools_count": 4 }, { "id": "british-transport", "name": "British Transport", "url": "/ai-mapping/organisations/british-transport/", "kml_url": "/ai-mapping", "tools_count": 1 }, { "id": "cambridgeshire", "name": "Cambridgeshire", "url": "/ai-mapping/organisations/cambridgeshire/", "kml_url": "/ai-mapping/assets/kml/cambridgeshire.kml", "tools_count": 4 }, { "id": "cheshire", "name": "Cheshire", "url": "/ai-mapping/organisations/cheshire/", "kml_url": "/ai-mapping/assets/kml/cheshire.kml", "tools_count": 1 }, { "id": "city-of-london", "name": "City of London", "url": "/ai-mapping/organisations/city-of-london/", "kml_url": "/ai-mapping/assets/kml/city-of-london.kml", "tools_count": 1 }, { "id": "cleveland", "name": "Cleveland", "url": "/ai-mapping/organisations/cleveland/", "kml_url": "/ai-mapping/assets/kml/cleveland.kml", "tools_count": 0 }, { "id": "cumbria", "name": "Cumbria", "url": "/ai-mapping/organisations/cumbria/", "kml_url": "/ai-mapping/assets/kml/cumbria.kml", "tools_count": 1 }, { "id": "derbyshire", "name": "Derbyshire", "url": "/ai-mapping/organisations/derbyshire/", "kml_url": "/ai-mapping/assets/kml/derbyshire.kml", "tools_count": 1 }, { "id": "devon-and-cornwall", "name": "Devon and Cornwall", "url": "/ai-mapping/organisations/devon-and-cornwall/", "kml_url": "/ai-mapping/assets/kml/devon-and-cornwall.kml", "tools_count": 2 }, { "id": "dorset", "name": "Dorset", "url": "/ai-mapping/organisations/dorset/", "kml_url": "/ai-mapping/assets/kml/dorset.kml", "tools_count": 0 }, { "id": "durham", "name": "Durham", "url": "/ai-mapping/organisations/durham/", "kml_url": "/ai-mapping/assets/kml/durham.kml", "tools_count": 0 }, { "id": "dyfed-powys", "name": "Dyfed Powys", "url": "/ai-mapping/organisations/dyfed-powys/", "kml_url": "/ai-mapping/assets/kml/dyfed-powys.kml", "tools_count": 0 }, { "id": "east-midlands-special-operations-unit", "name": "East Midlands Special Operations Unit", "url": "/ai-mapping/organisations/east-midlands-special-operations-unit/", "kml_url": "/ai-mapping", "tools_count": 0 }, { "id": "essex", "name": "Essex", "url": "/ai-mapping/organisations/essex/", "kml_url": "/ai-mapping/assets/kml/essex.kml", "tools_count": 6 }, { "id": "forensic-capability-network", "name": "Forensic Capability Network (FCN)", "url": "/ai-mapping/organisations/forensic-capability-network/", "kml_url": "/ai-mapping", "tools_count": 1 }, { "id": "gloucestershire", "name": "Gloucestershire", "url": "/ai-mapping/organisations/gloucestershire/", "kml_url": "/ai-mapping/assets/kml/gloucestershire.kml", "tools_count": 0 }, { "id": "greater-manchester", "name": "Greater Manchester", "url": "/ai-mapping/organisations/greater-manchester/", "kml_url": "/ai-mapping/assets/kml/greater-manchester.kml", "tools_count": 5 }, { "id": "gwent", "name": "Gwent", "url": "/ai-mapping/organisations/gwent/", "kml_url": "/ai-mapping/assets/kml/gwent.kml", "tools_count": 1 }, { "id": "hampshire", "name": "Hampshire", "url": "/ai-mapping/organisations/hampshire/", "kml_url": "/ai-mapping/assets/kml/hampshire.kml", "tools_count": 1 }, { "id": "hertfordshire", "name": "Hertfordshire", "url": "/ai-mapping/organisations/hertfordshire/", "kml_url": "/ai-mapping/assets/kml/hertfordshire.kml", "tools_count": 8 }, { "id": "home-office", "name": "Home Office", "url": "/ai-mapping/organisations/home-office/", "kml_url": "/ai-mapping", "tools_count": 1 }, { "id": "humberside", "name": "Humberside", "url": "/ai-mapping/organisations/humberside/", "kml_url": "/ai-mapping/assets/kml/humberside.kml", "tools_count": 4 }, { "id": "kent", "name": "Kent", "url": "/ai-mapping/organisations/kent/", "kml_url": "/ai-mapping/assets/kml/kent.kml", "tools_count": 4 }, { "id": "lancashire", "name": "Lancashire", "url": "/ai-mapping/organisations/lancashire/", "kml_url": "/ai-mapping/assets/kml/lancashire.kml", "tools_count": 1 }, { "id": "leicestershire", "name": "Leicestershire", "url": "/ai-mapping/organisations/leicestershire/", "kml_url": "/ai-mapping/assets/kml/leicestershire.kml", "tools_count": 1 }, { "id": "lincolnshire", "name": "Lincolnshire", "url": "/ai-mapping/organisations/lincolnshire/", "kml_url": "/ai-mapping/assets/kml/lincolnshire.kml", "tools_count": 2 }, { "id": "merseyside", "name": "Merseyside", "url": "/ai-mapping/organisations/merseyside/", "kml_url": "/ai-mapping/assets/kml/merseyside.kml", "tools_count": 4 }, { "id": "metropolitan-police", "name": "Metropolitan Police", "url": "/ai-mapping/organisations/metropolitan-police/", "kml_url": "/ai-mapping/assets/kml/metropolitan.kml", "tools_count": 9 }, { "id": "norfolk", "name": "Norfolk", "url": "/ai-mapping/organisations/norfolk/", "kml_url": "/ai-mapping/assets/kml/norfolk.kml", "tools_count": 3 }, { "id": "north-wales", "name": "North Wales", "url": "/ai-mapping/organisations/north-wales/", "kml_url": "/ai-mapping/assets/kml/north-wales.kml", "tools_count": 0 }, { "id": "north-yorkshire", "name": "North Yorkshire", "url": "/ai-mapping/organisations/north-yorkshire/", "kml_url": "/ai-mapping/assets/kml/north-yorkshire.kml", "tools_count": 0 }, { "id": "northamptonshire", "name": "Northamptonshire", "url": "/ai-mapping/organisations/northamptonshire/", "kml_url": "/ai-mapping/assets/kml/northamptonshire.kml", "tools_count": 2 }, { "id": "northern-ireland", "name": "Northern Ireland", "url": "/ai-mapping/organisations/northern-ireland/", "kml_url": "/ai-mapping", "tools_count": 0 }, { "id": "northumbria", "name": "Northumbria", "url": "/ai-mapping/organisations/northumbria/", "kml_url": "/ai-mapping/assets/kml/northumbria.kml", "tools_count": 1 }, { "id": "nottinghamshire", "name": "Nottinghamshire", "url": "/ai-mapping/organisations/nottinghamshire/", "kml_url": "/ai-mapping/assets/kml/nottinghamshire.kml", "tools_count": 2 }, { "id": "south-wales", "name": "South Wales", "url": "/ai-mapping/organisations/south-wales/", "kml_url": "/ai-mapping/assets/kml/south-wales.kml", "tools_count": 5 }, { "id": "south-yorkshire", "name": "South Yorkshire", "url": "/ai-mapping/organisations/south-yorkshire/", "kml_url": "/ai-mapping/assets/kml/south-yorkshire.kml", "tools_count": 1 }, { "id": "staffordshire", "name": "Staffordshire", "url": "/ai-mapping/organisations/staffordshire/", "kml_url": "/ai-mapping/assets/kml/staffordshire.kml", "tools_count": 1 }, { "id": "suffolk", "name": "Suffolk", "url": "/ai-mapping/organisations/suffolk/", "kml_url": "/ai-mapping/assets/kml/suffolk.kml", "tools_count": 4 }, { "id": "surrey", "name": "Surrey", "url": "/ai-mapping/organisations/surrey/", "kml_url": "/ai-mapping/assets/kml/surrey.kml", "tools_count": 3 }, { "id": "sussex", "name": "Sussex", "url": "/ai-mapping/organisations/sussex/", "kml_url": "/ai-mapping/assets/kml/sussex.kml", "tools_count": 3 }, { "id": "thames-valley", "name": "Thames Valley", "url": "/ai-mapping/organisations/thames-valley/", "kml_url": "/ai-mapping/assets/kml/thames-valley.kml", "tools_count": 8 }, { "id": "warwickshire", "name": "Warwickshire", "url": "/ai-mapping/organisations/warwickshire/", "kml_url": "/ai-mapping/assets/kml/warwickshire.kml", "tools_count": 0 }, { "id": "west-mercia", "name": "West Mercia", "url": "/ai-mapping/organisations/west-mercia/", "kml_url": "/ai-mapping/assets/kml/west-mercia.kml", "tools_count": 2 }, { "id": "west-midlands", "name": "West Midlands", "url": "/ai-mapping/organisations/west-midlands/", "kml_url": "/ai-mapping/assets/kml/west-midlands.kml", "tools_count": 16 }, { "id": "west-yorkshire", "name": "West Yorkshire", "url": "/ai-mapping/organisations/west-yorkshire/", "kml_url": "/ai-mapping/assets/kml/west-yorkshire.kml", "tools_count": 5 }, { "id": "wiltshire", "name": "Wiltshire", "url": "/ai-mapping/organisations/wiltshire/", "kml_url": "/ai-mapping/assets/kml/wiltshire.kml", "tools_count": 0 } ], "tools": [ { "id": "acusensus-ai-cameras", "title": "Acusensus AI cameras", "purpose": "Seatbelt/mobile phone use offences detection", "url": "/ai-mapping/tools/acusensus-ai-cameras/", "deployment_stage": "Trialled", "development_type": "Third-party", "developer_vendor": ["Acusensus","AECOM"], "users": ["Devon and Cornwall"], "criminal_justice_stages": [1,2,3], "inference_modes": ["Analysis"], "taxonomy": { "raw": ["Analysis(video -> [enum])"], "parsed": [ { "original": "Analysis(video -> [enum])", "inference_mode": "Analysis", "inputs": [ "video" ], "outputs": [ "[enum]" ] } ] }, "tool_functionality": ["seatbelt use and mobile phone distractions offences detection"], "resources": ["https://www.acusensus.com/devon-and-cornwall-police-commits-to-long-term-project-with-acusensus/","https://cornishstuff.com/2024/08/01/ai-cameras-crack-down-on-cornwalls-dangerous-drivers/"], "notes": null }, { "id": "ai-id-matching-system", "title": "AI ID Matching System", "purpose": "Remote check-in offender surveillance", "url": "/ai-mapping/tools/ai-id-matching-system/", "deployment_stage": "Trialled", "development_type": "Unknown", "developer_vendor": [], "users": ["Probation Service"], "criminal_justice_stages": [8], "inference_modes": ["Analysis"], "taxonomy": { "raw": ["Analysis(image, video -> enum)"], "parsed": [ { "original": "Analysis(image, video -> enum)", "inference_mode": "Analysis", "inputs": [ "image" ,"video" ], "outputs": [ "enum" ] } ] }, "tool_functionality": ["ID matching of person appearing in video and a reference image, where enum is boolean, i.e. yes/no if the ID of person is verified"], "resources": ["https://www.gov.uk/government/news/new-remote-face-scanning-tech-to-monitor-offenders-and-cut-crime"], "notes": "The pilot has been running across four probation regions in England; the South West, North West, East of England, and Kent, Surrey and Sussex." }, { "id": "ai-post-call-analysis", "title": "AI Post Call Analysis", "purpose": "Vulnerability detection", "url": "/ai-mapping/tools/ai-post-call-analysis/", "deployment_stage": "Trialled", "development_type": "In-house", "developer_vendor": ["Digital Public Contact (NPCC Contact portfolio)"], "users": ["Hertfordshire","West Yorkshire"], "criminal_justice_stages": [1,2,3], "inference_modes": ["Analysis","Generation"], "taxonomy": { "raw": ["Analysis(audio -> text)","Generation(prompt, text -> text)","Analysis(text -> enum)"], "parsed": [ { "original": "Analysis(audio -> text)", "inference_mode": "Analysis", "inputs": [ "audio" ], "outputs": [ "text" ] }, { "original": "Generation(prompt, text -> text)", "inference_mode": "Generation", "inputs": [ "prompt" ,"text" ], "outputs": [ "text" ] }, { "original": "Analysis(text -> enum)", "inference_mode": "Analysis", "inputs": [ "text" ], "outputs": [ "enum" ] } ] }, "tool_functionality": ["transcription","summarisation of call","categorisation of call by topic, repeat caller detection, vulnerability detection"], "resources": ["https://policinginsight.com/feature/innovation/ai-post-call-analysis-transforming-police-contact-insight/"], "notes": null }, { "id": "andi-esra", "title": "Andi-Esra", "purpose": "AI-assisted emergency call handling", "url": "/ai-mapping/tools/andi-esra/", "deployment_stage": "Live", "development_type": "Third-party", "developer_vendor": ["T-Tech"], "users": ["West Midlands"], "criminal_justice_stages": [1,2,3], "inference_modes": ["Analysis","Generation"], "taxonomy": { "raw": ["Analysis(audio -> text)","Generation(prompt, text, [enum] -> enum)"], "parsed": [ { "original": "Analysis(audio -> text)", "inference_mode": "Analysis", "inputs": [ "audio" ], "outputs": [ "text" ] }, { "original": "Generation(prompt, text, [enum] -> enum)", "inference_mode": "Generation", "inputs": [ "prompt" ,"text" ,"[enum]" ], "outputs": [ "enum" ] } ] }, "tool_functionality": ["transcription","search for keywords in transcript, where the first enum is the list of keywords, the second enum is boolean, i.e. yes/no to put the caller through a call handler"], "resources": ["https://www.westmidlands-pcc.gov.uk/pcc-praise-for-police-virtual-call-handling-assistant/","https://www.westmidlands-pcc.gov.uk/wp-content/uploads/2024/11/AGB-26.11.24-Agenda-Item-2-Minutes-of-Meeting.pdf?x48388"], "notes": null }, { "id": "arns", "title": "ARNS", "purpose": "Offenders' risk assessment", "url": "/ai-mapping/tools/arns/", "deployment_stage": "Trialled", "development_type": "Unknown", "developer_vendor": [], "users": ["Probation Service","Prison Service"], "criminal_justice_stages": [6,7,8], "inference_modes": ["Synthesis"], "taxonomy": { "raw": ["Synthesis(enum^n -> enum)"], "parsed": [ { "original": "Synthesis(enum^n -> enum)", "inference_mode": "Synthesis", "inputs": [ "enum^n" ], "outputs": [ "enum" ] } ] }, "tool_functionality": ["risk assessment"], "resources": ["https://www.judiciary.uk/wp-content/uploads/2024/08/2024-0409-Response-from-HMPPS-and-MoJ.pdf","https://committees.parliament.uk/publications/52323/documents/290569/default/","https://www.gov.uk/government/publications/ai-action-plan-for-justice/ai-action-plan-for-justice"], "notes": "There is limited information about the functionality of this tool, but its potential AI capabilities are reported." }, { "id": "auror", "title": "Auror", "purpose": "Match offenders across UK stores", "url": "/ai-mapping/tools/auror/", "deployment_stage": "Trialled", "development_type": "Third-party", "developer_vendor": ["Auror"], "users": ["Hertfordshire","Devon and Cornwall","City of London","Nottinghamshire"], "criminal_justice_stages": [1,2,3], "inference_modes": ["Analysis"], "taxonomy": { "raw": ["Analysis(video, [image] -> [enum])"], "parsed": [ { "original": "Analysis(video, [image] -> [enum])", "inference_mode": "Analysis", "inputs": [ "video" ,"[image]" ], "outputs": [ "[enum]" ] } ] }, "tool_functionality": ["face recognition"], "resources": ["https://www.auror.co/uk","https://news.devon-cornwall.police.uk/news-article/fe465a3e-c5f9-ef11-9d75-6045bdd24049#:~:text=Devon%20and%20Cornwall%20has%20recently,crime%20in%20Devon%20and%20Cornwall.","https://democracy.cityoflondon.gov.uk/documents/s218378/Productivity%20action%20plan%20for%20RREC.pdf","https://www.nottinghamshire.police.uk/news/nottinghamshire/news/news/2025/july/force-pilots-use-of-crime-intelligence-platform-to-help-combat-organised-retail-crime/"], "notes": null }, { "id": "auto-transcription-and-summarisation-tool", "title": "Auto transcription and summarisation tool", "purpose": "Call summarisation", "url": "/ai-mapping/tools/auto-transcription-and-summarisation-tool/", "deployment_stage": "Trialled", "development_type": "Unknown", "developer_vendor": [], "users": ["Avon and Somerset"], "criminal_justice_stages": [1,2,3], "inference_modes": ["Analysis","Generation"], "taxonomy": { "raw": ["Analysis(audio -> text)","Generation(prompt, text -> text)"], "parsed": [ { "original": "Analysis(audio -> text)", "inference_mode": "Analysis", "inputs": [ "audio" ], "outputs": [ "text" ] }, { "original": "Generation(prompt, text -> text)", "inference_mode": "Generation", "inputs": [ "prompt" ,"text" ], "outputs": [ "text" ] } ] }, "tool_functionality": ["transcription","summarisation"], "resources": [], "notes": null }, { "id": "bobbi", "title": "Bobbi", "purpose": "AI webchat for non-emergency public queries", "url": "/ai-mapping/tools/bobbi/", "deployment_stage": "Live", "development_type": "Third-party", "developer_vendor": ["Salesforce"], "users": ["Thames Valley","Hampshire and Isle of Wight","Humberside","Staffordshire"], "criminal_justice_stages": [1], "inference_modes": ["Generation","Analysis"], "taxonomy": { "raw": ["Generation(prompt, enum -> text, [enum])","Analysis(text -> enum)"], "parsed": [ { "original": "Generation(prompt, enum -> text, [enum])", "inference_mode": "Generation", "inputs": [ "prompt" ,"enum" ], "outputs": [ "text" ,"[enum]" ] }, { "original": "Analysis(text -> enum)", "inference_mode": "Analysis", "inputs": [ "text" ], "outputs": [ "enum" ] } ] }, "tool_functionality": ["generation of a text response to a query from the public, with a list of citations referring to force’s knowledge articles – including policy documents and public-facing guidance, given a prompt and named repository. Here, 1st enum is repository names and 2nd is citations (an example of RAG - retrieval augmented generation)","vulnerability detection by identifying markers of vulnerability, where enum is boolean, i.e. yes/no"], "resources": ["https://www.thamesvalley.police.uk/news/thames-valley/news/2025/11-november/24-11-2025/meet-bobbi-policings-new-ai-virtual-assistant/","https://www.bbc.co.uk/news/articles/cdxw9xnk7rzo","https://www.salesforce.com/uk/news/press-releases/2025/12/04/uk-police-forces-agentforce/","https://www.bbc.co.uk/news/articles/cvgq03y0l3yo","https://www.salesforce.com/uk/agentforce/"], "notes": null }, { "id": "briefcam", "title": "BriefCam", "purpose": "Video footage search", "url": "/ai-mapping/tools/briefcam/", "deployment_stage": "Live", "development_type": "Third-party", "developer_vendor": ["BriefCam"], "users": ["West Midlands"], "criminal_justice_stages": [3], "inference_modes": ["Analysis"], "taxonomy": { "raw": ["Analysis(video, [image] -> [enum])"], "parsed": [ { "original": "Analysis(video, [image] -> [enum])", "inference_mode": "Analysis", "inputs": [ "video" ,"[image]" ], "outputs": [ "[enum]" ] } ] }, "tool_functionality": ["face and object recognition"], "resources": ["https://www.briefcam.com/"], "notes": null }, { "id": "casepilot", "title": "CasePilot", "purpose": "AI-assisted police case checker", "url": "/ai-mapping/tools/casepilot/", "deployment_stage": "Trialled", "development_type": "Third-party", "developer_vendor": ["Fuzzy Labs","Sheffield Hallam University","Microsoft"], "users": ["South Yorkshire"], "criminal_justice_stages": [3], "inference_modes": ["Generation"], "taxonomy": { "raw": ["Generation(prompt, text, [enum] -> [enum, text])","Generation(prompt, text -> text)","Generation(prompt, text -> [enum])","Generation(prompt, [text] -> [enum])"], "parsed": [ { "original": "Generation(prompt, text, [enum] -> [enum, text])", "inference_mode": "Generation", "inputs": [ "prompt" ,"text" ,"[enum]" ], "outputs": [ "[enum" ,"text]" ] }, { "original": "Generation(prompt, text -> text)", "inference_mode": "Generation", "inputs": [ "prompt" ,"text" ], "outputs": [ "text" ] }, { "original": "Generation(prompt, text -> [enum])", "inference_mode": "Generation", "inputs": [ "prompt" ,"text" ], "outputs": [ "[enum]" ] }, { "original": "Generation(prompt, [text] -> [enum])", "inference_mode": "Generation", "inputs": [ "prompt" ,"[text]" ], "outputs": [ "[enum]" ] } ] }, "tool_functionality": ["assessment of evidence against points to prove","summarisation","identifying individuals involved as part of the case and their role; key events as part of the case and detailed timelines","corroborations and discrepancies of interviews and other key evidence"], "resources": ["https://www.npcc.police.uk/SysSiteAssets/media/downloads/publications/disclosure-logs/npcc-central-office/2024/140-2024-police-star-fund-projects-24-25---april-amended.pdf","https://science.police.uk/opportunities/police-star-fund/previously-funded-police-star-fund-projects/","https://www.fuzzylabs.ai/case-studies/building-an-ai-assistant-to-halve-police-case-file-prep-time"], "notes": null }, { "id": "cctv-analysis-in-prisons", "title": "CCTV analysis in Prisons", "purpose": "Protection of staff and individuals in custody", "url": "/ai-mapping/tools/cctv-analysis-in-prisons/", "deployment_stage": "Experimental", "development_type": "Unknown", "developer_vendor": [], "users": ["Prison Service"], "criminal_justice_stages": [7], "inference_modes": ["Analysis"], "taxonomy": { "raw": ["Analysis(video -> enum)"], "parsed": [ { "original": "Analysis(video -> enum)", "inference_mode": "Analysis", "inputs": [ "video" ], "outputs": [ "enum" ] } ] }, "tool_functionality": ["detection of safety-critical incidents in custodial settings"], "resources": ["https://ai.justice.gov.uk/our-work/cctv"], "notes": null }, { "id": "cctv-system-equipped-with-artificial-intelligence-analytics", "title": "CCTV system equipped with artificial intelligence analytics", "purpose": "ID missing people/vehicles, suspects", "url": "/ai-mapping/tools/cctv-system-equipped-with-artificial-intelligence-analytics/", "deployment_stage": "Live", "development_type": "Unknown", "developer_vendor": [], "users": ["Cumbria","Metropolitan Police"], "criminal_justice_stages": [2,3], "inference_modes": ["Analysis"], "taxonomy": { "raw": ["Analysis(video, enum -> [enum])","Analysis(video -> [enum])"], "parsed": [ { "original": "Analysis(video, enum -> [enum])", "inference_mode": "Analysis", "inputs": [ "video" ,"enum" ], "outputs": [ "[enum]" ] }, { "original": "Analysis(video -> [enum])", "inference_mode": "Analysis", "inputs": [ "video" ], "outputs": [ "[enum]" ] } ] }, "tool_functionality": ["recognition of hats, face coverings, backpacks and whether anything is carried in hand, such as a bag or umbrella; first enum refers to a search filter (e.g. blue top), second enum refers to indices to video frames where the recognition was achieved","face detection, prediction of sex (F/M), age (adult/child)"], "resources": ["https://policinginsight.com/feature/cumbrias-ai-enabled-cctv-saving-lives-and-thousands-of-police-hours/","https://www.news.cumbria.police.uk/news/cctv-analytical-capability-is-helping-police-catch-criminals-and-safeguard-the-vulnerable-in-a-fraction-of-the-time","https://www.youtube.com/watch?v=lTHIV--wJzU","https://www.biometricupdate.com/202509/london-police-target-video-analytics-as-part-of-25m-digital-market-strategy"], "notes": "Metropolitan Police is trialling this capability." }, { "id": "cecil", "title": "Cecil", "purpose": "Officer guidance", "url": "/ai-mapping/tools/cecil/", "deployment_stage": "Trialled", "development_type": "Third-party", "developer_vendor": ["Coeus Software"], "users": ["Humberside"], "criminal_justice_stages": [1,3], "inference_modes": ["Generation"], "taxonomy": { "raw": ["Generation(prompt, enum -> text, [enum])"], "parsed": [ { "original": "Generation(prompt, enum -> text, [enum])", "inference_mode": "Generation", "inputs": [ "prompt" ,"enum" ], "outputs": [ "text" ,"[enum]" ] } ] }, "tool_functionality": ["generation of a text response, with a list of citations, given a prompt and named repository. Here, 1st enum is repository names and 2nd is citations (an example of RAG - retrieval augmented generation)"], "resources": ["https://www.uktech.news/ai/yorkshire-police-force-declares-successful-ai-trial-results-20251105"], "notes": null }, { "id": "cesium", "title": "CESIUM", "purpose": "Children at risk of exploitation identification", "url": "/ai-mapping/tools/cesium/", "deployment_stage": "Live", "development_type": "Third-party", "developer_vendor": ["Trilateral Research","Lincolnshire"], "users": ["Lincolnshire"], "criminal_justice_stages": [1], "inference_modes": ["Synthesis","Analysis"], "taxonomy": { "raw": ["Synthesis(enum^n -> enum)","Analysis(text -> enum)"], "parsed": [ { "original": "Synthesis(enum^n -> enum)", "inference_mode": "Synthesis", "inputs": [ "enum^n" ], "outputs": [ "enum" ] }, { "original": "Analysis(text -> enum)", "inference_mode": "Analysis", "inputs": [ "text" ], "outputs": [ "enum" ] } ] }, "tool_functionality": ["risk assessment","classify and code statements"], "resources": ["https://www.ukauthority.com/articles/lincolnshire-police-deploys-ai-for-child-safeguarding/","https://lincolnshire-pcc.gov.uk/wp-content/uploads/2024/04/24-2023-pcc-decision-paper-cesium-v2-web-version.pdf","https://trilateralresearch.com/cesium-application","https://trilateralresearch.com/cesium-case-study"], "notes": null }, { "id": "chatgpt-enterprise", "title": "ChatGPT Enterprise", "purpose": "Summarisation and report writing", "url": "/ai-mapping/tools/chatgpt-enterprise/", "deployment_stage": "Live", "development_type": "Third-party", "developer_vendor": ["OpenAI"], "users": ["Ministry of Justice"], "criminal_justice_stages": [4,5], "inference_modes": ["Generation"], "taxonomy": { "raw": ["Generation(prompt, text -> text)"], "parsed": [ { "original": "Generation(prompt, text -> text)", "inference_mode": "Generation", "inputs": [ "prompt" ,"text" ], "outputs": [ "text" ] } ] }, "tool_functionality": ["summarisation, report writing"], "resources": ["https://www.gov.uk/government/publications/ai-action-plan-for-justice/ai-action-plan-for-justice#strengthen-our-foundations","https://ai.justice.gov.uk/our-work/chatgpt-enterprise"], "notes": null }, { "id": "cocounsel", "title": "CoCounsel", "purpose": "AI-assisted research in Westlaw UK and Practical Law UK", "url": "/ai-mapping/tools/cocounsel/", "deployment_stage": "Live", "development_type": "Third-party", "developer_vendor": ["Thomson Reuters"], "users": ["Legal professionals"], "criminal_justice_stages": [3,4,5], "inference_modes": ["Generation"], "taxonomy": { "raw": ["Generation(prompt, enum -> text, [enum])"], "parsed": [ { "original": "Generation(prompt, enum -> text, [enum])", "inference_mode": "Generation", "inputs": [ "prompt" ,"enum" ], "outputs": [ "text" ,"[enum]" ] } ] }, "tool_functionality": ["generation of a text response, with a list of citations referring to Westlaw UK or Practical Law UK repositories, given a prompt and named repository. Here, 1st enum is repository names and 2nd is citations (an example of RAG - retrieval augmented generation)"], "resources": ["https://legalsolutions.thomsonreuters.co.uk/en/c/cocounsel-legal-uk/one-comprehensive-solution-built-for-legal-professionals.html?searchid=TRPPCSOL/Google/LegalUK_RS_CoCounsel_Search_Brand-All_UK/CoCounsel"], "notes": null }, { "id": "collaboraite-suite", "title": "Collaboraite Suite", "purpose": "AI pipelines building", "url": "/ai-mapping/tools/collaboraite-suite/", "deployment_stage": "Stage Unknown", "development_type": "Third-party", "developer_vendor": ["Collaboraite"], "users": ["West Midlands"], "criminal_justice_stages": [1,2,3], "inference_modes": ["Analysis","Analaysis","Generation"], "taxonomy": { "raw": ["Analysis(text, [enum] -> [enum, enum])","Analaysis(text -> text)","Analysis(audio -> text)","Generation(prompt, video, text -> [enum])"], "parsed": [ { "original": "Analysis(text, [enum] -> [enum, enum])", "inference_mode": "Analysis", "inputs": [ "text" ,"[enum]" ], "outputs": [ "[enum" ,"enum]" ] }, { "original": "Analaysis(text -> text)", "inference_mode": "Analaysis", "inputs": [ "text" ], "outputs": [ "text" ] }, { "original": "Analysis(audio -> text)", "inference_mode": "Analysis", "inputs": [ "audio" ], "outputs": [ "text" ] }, { "original": "Generation(prompt, video, text -> [enum])", "inference_mode": "Generation", "inputs": [ "prompt" ,"video" ,"text" ], "outputs": [ "[enum]" ] } ] }, "tool_functionality": ["search for keywords in text, where first enum is the keyword list, second enum is a keyword from the list, and third enum is the index (i.e. position) in text","translation","transcription","object recognition in video using natural language (text) queries, where enum is the index in video"], "resources": ["https://collaboraite.co.uk/","https://assets.applytosupply.digitalmarketplace.service.gov.uk/g-cloud-14/documents/702997/729871546736421-service-definition-document-2024-12-23-1533.pdf"], "notes": null }, { "id": "copa", "title": "CoPA", "purpose": "Officer guidance", "url": "/ai-mapping/tools/copa/", "deployment_stage": "Live", "development_type": "In-house", "developer_vendor": ["British Transport"], "users": ["British Transport","Greater Manchester"], "criminal_justice_stages": [1,3], "inference_modes": ["Generation"], "taxonomy": { "raw": ["Generation(prompt, enum -> text, [enum])"], "parsed": [ { "original": "Generation(prompt, enum -> text, [enum])", "inference_mode": "Generation", "inputs": [ "prompt" ,"enum" ], "outputs": [ "text" ,"[enum]" ] } ] }, "tool_functionality": ["generation of a text response, with a list of citations, given a prompt and named repository. Here, 1st enum is repository names and 2nd is citations (an example of RAG - retrieval augmented generation)"], "resources": [], "notes": null }, { "id": "copilot", "title": "Copilot", "purpose": "Summarisation, report writing & agentic task coordination", "url": "/ai-mapping/tools/copilot/", "deployment_stage": "Live", "development_type": "Third-party", "developer_vendor": ["Microsoft"], "users": ["Kent","Metropolitan Police","West Midlands","Hertfordshire","Thames Valley","Hampshire and Isle of Wight","Home Office","Crown Prosecution Service","Probation Service","Ministry of Justice"], "criminal_justice_stages": [1,2,3,4,5,8], "inference_modes": ["Generation"], "taxonomy": { "raw": ["Generation(prompt, text -> text)"], "parsed": [ { "original": "Generation(prompt, text -> text)", "inference_mode": "Generation", "inputs": [ "prompt" ,"text" ], "outputs": [ "text" ] } ] }, "tool_functionality": ["summarisation, report writing"], "resources": [], "notes": "Home Office has deployed this, the rest of users have trialled it. Copilot is live in all 43 forces across England and Wales and it's up to each kind of force to determine whether or not they turn on that functionality." }, { "id": "correspondence-drafting-tool", "title": "Correspondence Drafting Tool", "purpose": "Correspondence drafting", "url": "/ai-mapping/tools/correspondence-drafting-tool/", "deployment_stage": "Live", "development_type": "Third-party", "developer_vendor": ["NTT Data UK Limited"], "users": ["Crown Prosecution Service"], "criminal_justice_stages": [4,5], "inference_modes": ["Generation"], "taxonomy": { "raw": ["Generation(prompt, text -> text)"], "parsed": [ { "original": "Generation(prompt, text -> text)", "inference_mode": "Generation", "inputs": [ "prompt" ,"text" ], "outputs": [ "text" ] } ] }, "tool_functionality": ["generation of first drafts of letters and emails to victims by pre-populating mandatory information through extracting information from the CPS case management system; summarisation"], "resources": ["https://www.gov.uk/algorithmic-transparency-records/the-crown-prosecution-service-correspondence-drafting-tool","https://www.government-transformation.com/data/cps-trials-ai-tool-to-transform-routine-correspondence"], "notes": null }, { "id": "corvus-iom-case", "title": "Corvus IOM Case", "purpose": "Offender reoffending risk assessment", "url": "/ai-mapping/tools/corvus-iom-case/", "deployment_stage": "Live", "development_type": "Third-party", "developer_vendor": ["Bluestar"], "users": ["West Yorkshire"], "criminal_justice_stages": [1,2,7,8], "inference_modes": ["Synthesis"], "taxonomy": { "raw": ["Synthesis(enum^n -> enum)"], "parsed": [ { "original": "Synthesis(enum^n -> enum)", "inference_mode": "Synthesis", "inputs": [ "enum^n" ], "outputs": [ "enum" ] } ] }, "tool_functionality": ["offence risk profiling"], "resources": ["https://www.westyorkshire.police.uk/sites/default/files/2023-09/integrated_offender_management_iom.pdf","https://bluestar-software.co.uk/products/the-corvus-platform/"], "notes": null }, { "id": "definely-suite", "title": "Definely Suite", "purpose": "Legal documents drafting", "url": "/ai-mapping/tools/definely-suite/", "deployment_stage": "Live", "development_type": "Third-party", "developer_vendor": [], "users": ["Legal professionals"], "criminal_justice_stages": [3,4,5], "inference_modes": ["Analysis","Generation"], "taxonomy": { "raw": ["Analysis([text] -> [text, enum])","Generation(prompt, text -> [text, enum])","Generation(prompt, text -> text)","Generation(prompt, text, [enum] -> text)","Generation(prompt, text, text -> [text, enum])"], "parsed": [ { "original": "Analysis([text] -> [text, enum])", "inference_mode": "Analysis", "inputs": [ "[text]" ], "outputs": [ "[text" ,"enum]" ] }, { "original": "Generation(prompt, text -> [text, enum])", "inference_mode": "Generation", "inputs": [ "prompt" ,"text" ], "outputs": [ "[text" ,"enum]" ] }, { "original": "Generation(prompt, text -> text)", "inference_mode": "Generation", "inputs": [ "prompt" ,"text" ], "outputs": [ "text" ] }, { "original": "Generation(prompt, text, [enum] -> text)", "inference_mode": "Generation", "inputs": [ "prompt" ,"text" ,"[enum]" ], "outputs": [ "text" ] }, { "original": "Generation(prompt, text, text -> [text, enum])", "inference_mode": "Generation", "inputs": [ "prompt" ,"text" ,"text" ], "outputs": [ "[text" ,"enum]" ] } ] }, "tool_functionality": ["categorisation of clauses and definitions in the organisation's documents","retrieval of key information from documents","redrafting of documents given some editing guidelines","summarisation of a user-defined list of clauses, redaction of document given a list of information to be redacted","contract analysis that tracks the knock-on effects of every change"], "resources": ["https://www.definely.com/"], "notes": null }, { "id": "domestic-abuse-risk-assessment-tool-darat", "title": "Domestic Abuse Risk Assessment Tool (DARAT)", "purpose": "Future domestic harm risk assessment", "url": "/ai-mapping/tools/domestic-abuse-risk-assessment-tool-darat/", "deployment_stage": "Experimental", "development_type": "Third-party", "developer_vendor": ["Solvarithm"], "users": ["Hampshire and Isle of Wight","Thames Valley"], "criminal_justice_stages": [1,2], "inference_modes": ["Synthesis"], "taxonomy": { "raw": ["Synthesis(enum^n -> enum)"], "parsed": [ { "original": "Synthesis(enum^n -> enum)", "inference_mode": "Synthesis", "inputs": [ "enum^n" ], "outputs": [ "enum" ] } ] }, "tool_functionality": ["future domestic harm risk profiling"], "resources": ["https://www.gov.uk/government/publications/hampshire-and-thames-valley-police-darat","https://www.college.police.uk/article/police-better-equipped-spot-controlling-behaviour","https://www.college.police.uk/article/police-better-equipped-spot-controlling-behaviour","https://assets.college.police.uk/s3fs-public/2022-12/CoP-Domestic%20Abuse%20Risk%20Assessment.pdf","https://www.gov.uk/algorithmic-transparency-records/hampshire-and-thames-valley-police-darat"], "notes": null }, { "id": "dragon-spotter", "title": "DRAGON-Spotter", "purpose": "ID grooming behaviour", "url": "/ai-mapping/tools/dragon-spotter/", "deployment_stage": "Trialled", "development_type": "Academic collaboration", "developer_vendor": ["Swansea University"], "users": ["West Mercia"], "criminal_justice_stages": [1,2,3], "inference_modes": ["Analysis"], "taxonomy": { "raw": ["Analysis(text -> enum)"], "parsed": [ { "original": "Analysis(text -> enum)", "inference_mode": "Analysis", "inputs": [ "text" ], "outputs": [ "enum" ] } ] }, "tool_functionality": ["identify grooming behaviour"], "resources": ["https://www.npcc.police.uk/SysSiteAssets/media/downloads/publications/disclosure-logs/npcc-central-office/2024/140-2024-police-star-fund-projects-24-25---april-amended.pdf","https://science.police.uk/site/assets/files/3795/revolutionising_digital_forensics_child_sexual_exploitation_workflows_through_linguistics_ai_technology_dragon-spotter.pdf"], "notes": null }, { "id": "enquiry-calls-analysis", "title": "Enquiry calls analysis", "purpose": "Recurring patterns identification in enquiry calls", "url": "/ai-mapping/tools/enquiry-calls-analysis/", "deployment_stage": "Trialled", "development_type": "In-house", "developer_vendor": ["Merseyside"], "users": ["Merseyside"], "criminal_justice_stages": [1,2], "inference_modes": ["Analysis","Generation"], "taxonomy": { "raw": ["Analysis(audio -> text)","Generation(prompt, text -> text)","Generation(prompt, [text] -> [enum])"], "parsed": [ { "original": "Analysis(audio -> text)", "inference_mode": "Analysis", "inputs": [ "audio" ], "outputs": [ "text" ] }, { "original": "Generation(prompt, text -> text)", "inference_mode": "Generation", "inputs": [ "prompt" ,"text" ], "outputs": [ "text" ] }, { "original": "Generation(prompt, [text] -> [enum])", "inference_mode": "Generation", "inputs": [ "prompt" ,"[text]" ], "outputs": [ "[enum]" ] } ] }, "tool_functionality": ["transcription","summarisation","topic modelling of a call transcript (generates topics from text)"], "resources": [], "notes": null }, { "id": "equip", "title": "EQuiP", "purpose": "Policies and practice guidance search and retrieval", "url": "/ai-mapping/tools/equip/", "deployment_stage": "Experimental", "development_type": "Unknown", "developer_vendor": [], "users": ["Probation Service"], "criminal_justice_stages": [8], "inference_modes": ["Generation"], "taxonomy": { "raw": ["Generation(prompt, enum -> text, [enum])"], "parsed": [ { "original": "Generation(prompt, enum -> text, [enum])", "inference_mode": "Generation", "inputs": [ "prompt" ,"enum" ], "outputs": [ "text" ,"[enum]" ] } ] }, "tool_functionality": ["generation of a text response, with a list of citations, given a prompt and named repository. Here, 1st enum is repository names and 2nd is citations (an example of RAG - retrieval augmented generation)"], "resources": ["https://hmiprobation.justiceinspectorates.gov.uk/document/national-inspection-april-2025/"], "notes": null }, { "id": "evidence-based-investigative-tool-ebit", "title": "Evidence-Based Investigative Tool (EBIT)", "purpose": "Crime solvability assessment", "url": "/ai-mapping/tools/evidence-based-investigative-tool-ebit/", "deployment_stage": "Live", "development_type": "In-house", "developer_vendor": ["Kent"], "users": ["Kent"], "criminal_justice_stages": [3], "inference_modes": ["Synthesis"], "taxonomy": { "raw": ["Synthesis(enum^n -> enum)"], "parsed": [ { "original": "Synthesis(enum^n -> enum)", "inference_mode": "Synthesis", "inputs": [ "enum^n" ], "outputs": [ "enum" ] } ] }, "tool_functionality": ["predict investigative success (further investigation, close the case pending further evidence, or further review by a supervisor)"], "resources": ["https://www.kent.police.uk/SysSiteAssets/foi-media/kent/how-we-make-decisions/force-management-statement/redacted-kent-police-fms-2024--.pdf"], "notes": null }, { "id": "evidence-com", "title": "Evidence.com", "purpose": "Body-worn camera videos analysis", "url": "/ai-mapping/tools/evidence-com/", "deployment_stage": "Live", "development_type": "Third-party", "developer_vendor": ["BriefCam"], "users": ["West Midlands"], "criminal_justice_stages": [1,2,3], "inference_modes": ["Analysis","Generation"], "taxonomy": { "raw": ["Analysis(audio -> text)","Analysis(video, image -> [enum])","Generation(prompt, text -> text)"], "parsed": [ { "original": "Analysis(audio -> text)", "inference_mode": "Analysis", "inputs": [ "audio" ], "outputs": [ "text" ] }, { "original": "Analysis(video, image -> [enum])", "inference_mode": "Analysis", "inputs": [ "video" ,"image" ], "outputs": [ "[enum]" ] }, { "original": "Generation(prompt, text -> text)", "inference_mode": "Generation", "inputs": [ "prompt" ,"text" ], "outputs": [ "text" ] } ] }, "tool_functionality": ["transcription","object recognition","summarisation, document generation"], "resources": ["https://www.westmidlands.police.uk/SysSiteAssets/foi-media/west-midlands/disclosure_log_2024/august/body-worn-video-v1.0.pdf"], "notes": null }, { "id": "extraction-of-missing-modus-operandi-mo-data-from-crime-records", "title": "Extraction of missing Modus Operandi (MO) data from crime records", "purpose": "Missing Modus Operandi (MO) data extraction from crime records", "url": "/ai-mapping/tools/extraction-of-missing-modus-operandi-mo-data-from-crime-records/", "deployment_stage": "Trialled", "development_type": "Unknown", "developer_vendor": [], "users": ["Merseyside"], "criminal_justice_stages": [1,2,3], "inference_modes": ["Generation"], "taxonomy": { "raw": ["Generation(prompt, [text] -> text)"], "parsed": [ { "original": "Generation(prompt, [text] -> text)", "inference_mode": "Generation", "inputs": [ "prompt" ,"[text]" ], "outputs": [ "text" ] } ] }, "tool_functionality": ["summarisation; generation of outputs for MO fields"], "resources": [], "notes": null }, { "id": "genai-platform", "title": "GenAI Platform", "purpose": "Calls and crime categorisation", "url": "/ai-mapping/tools/genai-platform/", "deployment_stage": "Trialled", "development_type": "Third-party", "developer_vendor": ["WatsonX IBM"], "users": ["Avon and Somerset"], "criminal_justice_stages": [1,2,3], "inference_modes": ["Analysis","Generation"], "taxonomy": { "raw": ["Analysis(audio -> text)","Generation(text, prompt -> text)","Analysis(text -> enum)"], "parsed": [ { "original": "Analysis(audio -> text)", "inference_mode": "Analysis", "inputs": [ "audio" ], "outputs": [ "text" ] }, { "original": "Generation(text, prompt -> text)", "inference_mode": "Generation", "inputs": [ "text" ,"prompt" ], "outputs": [ "text" ] }, { "original": "Analysis(text -> enum)", "inference_mode": "Analysis", "inputs": [ "text" ], "outputs": [ "enum" ] } ] }, "tool_functionality": ["transcription","summarisation","call categorisation, crime classification"], "resources": [], "notes": null }, { "id": "goodsam-with-ai-capabilities", "title": "GoodSAM (with AI capabilities)", "purpose": "Transcription & document generation-enhanced video conferencing", "url": "/ai-mapping/tools/goodsam-with-ai-capabilities/", "deployment_stage": "Trialled", "development_type": "Third-party", "developer_vendor": ["GoodSAM"], "users": ["Essex","West Midlands"], "criminal_justice_stages": [3,4,5], "inference_modes": ["Analysis","Generation"], "taxonomy": { "raw": ["Analysis(audio -> text)","Generation(prompt, text -> text)"], "parsed": [ { "original": "Analysis(audio -> text)", "inference_mode": "Analysis", "inputs": [ "audio" ], "outputs": [ "text" ] }, { "original": "Generation(prompt, text -> text)", "inference_mode": "Generation", "inputs": [ "prompt" ,"text" ], "outputs": [ "text" ] } ] }, "tool_functionality": ["transcription","witness statements writing"], "resources": ["https://www.goodsamapp.org/police"], "notes": null }, { "id": "humphrey-suite", "title": "Humphrey Suite", "purpose": "UK civil servants AI assistance", "url": "/ai-mapping/tools/humphrey-suite/", "deployment_stage": "Live", "development_type": "In-house", "developer_vendor": ["i.AI (UK Government incubator for AI)"], "users": ["Crown Prosecution Service","England and Wales Civil Service"], "criminal_justice_stages": [4,5], "inference_modes": ["Analysis","Generation"], "taxonomy": { "raw": ["Analysis(audio -> text)","Generation(prompt, text -> text)","Analysis([text] -> [enum, enum])"], "parsed": [ { "original": "Analysis(audio -> text)", "inference_mode": "Analysis", "inputs": [ "audio" ], "outputs": [ "text" ] }, { "original": "Generation(prompt, text -> text)", "inference_mode": "Generation", "inputs": [ "prompt" ,"text" ], "outputs": [ "text" ] }, { "original": "Analysis([text] -> [enum, enum])", "inference_mode": "Analysis", "inputs": [ "[text]" ], "outputs": [ "[enum" ,"enum]" ] } ] }, "tool_functionality": ["transcription","summarisation","extraction of patterns and themes from the public consultation responses"], "resources": ["https://www.gov.uk/government/news/government-built-humphrey-ai-tool-reviews-responses-to-consultation-for-first-time-in-bid-to-save-millions","https://www.globalgovernmentforum.com/yes-civil-servant-meet-humphrey-the-governments-ai-package-for-officials/","https://www.gov.uk/government/news/ai-experiments-see-humphrey-help-townhalls-cut-costs-and-improve-services","https://github.com/i-dot-ai"], "notes": null }, { "id": "ihotspot", "title": "iHotSpot", "purpose": "Daily crime hotspot prediction", "url": "/ai-mapping/tools/ihotspot/", "deployment_stage": "Live", "development_type": "Third-party", "developer_vendor": ["SpaceTimeAI"], "users": ["Metropolitan Police"], "criminal_justice_stages": [1], "inference_modes": ["Synthesis"], "taxonomy": { "raw": ["Synthesis(enum^n -> enum)"], "parsed": [ { "original": "Synthesis(enum^n -> enum)", "inference_mode": "Synthesis", "inputs": [ "enum^n" ], "outputs": [ "enum" ] } ] }, "tool_functionality": ["crime hot spot"], "resources": [], "notes": null }, { "id": "integrated-offender-management-iom-model", "title": "Integrated Offender Management (IOM) model", "purpose": "Predictive offender risk assessment", "url": "/ai-mapping/tools/integrated-offender-management-iom-model/", "deployment_stage": "Trialled", "development_type": "In-house", "developer_vendor": ["West Midlands"], "users": ["West Midlands"], "criminal_justice_stages": [1,2], "inference_modes": ["Synthesis"], "taxonomy": { "raw": ["Synthesis(enum^n -> enum)"], "parsed": [ { "original": "Synthesis(enum^n -> enum)", "inference_mode": "Synthesis", "inputs": [ "enum^n" ], "outputs": [ "enum" ] } ] }, "tool_functionality": ["offence risk profiling"], "resources": ["https://foi.west-midlands.police.uk/wp-content/uploads/2022/08/855A_ATTACHMENT_01.pdf","https://www.westmidlands-pcc.gov.uk/wp-content/uploads/2023/01/3.1-Internal-RFSDi-IOM-Evaluation-v1.0-1-1.pdf?x59042"], "notes": null }, { "id": "judicial-case-triage-and-listing-tool", "title": "Judicial case triage and listing tool", "purpose": "Trial-ready cases identification", "url": "/ai-mapping/tools/judicial-case-triage-and-listing-tool/", "deployment_stage": "Experimental", "development_type": "Unknown", "developer_vendor": [], "users": ["Courts"], "criminal_justice_stages": [5], "inference_modes": [], "taxonomy": { "raw": [], "parsed": [ ] }, "tool_functionality": [], "resources": ["https://www.gov.uk/government/news/ai-tech-ambition-to-deliver-smarter-justice-for-victims"], "notes": "Used by Crown Courts. Due to limited information, we are unable to map out the taxonomy." }, { "id": "justice-transcribe", "title": "Justice Transcribe", "purpose": "AI transcription and summarisation tools", "url": "/ai-mapping/tools/justice-transcribe/", "deployment_stage": "Live", "development_type": "In-house", "developer_vendor": ["Justice AI Unit"], "users": ["Probation Service","Courts"], "criminal_justice_stages": [4,5,8], "inference_modes": ["Analysis","Generation"], "taxonomy": { "raw": ["Analysis(audio -> text)","Generation(prompt, text -> text)"], "parsed": [ { "original": "Analysis(audio -> text)", "inference_mode": "Analysis", "inputs": [ "audio" ], "outputs": [ "text" ] }, { "original": "Generation(prompt, text -> text)", "inference_mode": "Generation", "inputs": [ "prompt" ,"text" ], "outputs": [ "text" ] } ] }, "tool_functionality": ["transcription","summarisation"], "resources": ["https://assets.publishing.service.gov.uk/media/699c4fac31713b50fd49c033/Justice-transcribe-report.pdf","https://www.gov.uk/government/publications/ai-action-plan-for-justice/ai-action-plan-for-justice","https://www.gov.uk/government/news/ai-tech-ambition-to-deliver-smarter-justice-for-victims","https://ai.justice.gov.uk/our-work/justice-transcribe-in-the-courts"], "notes": "It was first piloted in Kent, Surrey, Sussex, and Wales probation services but has now been scaled across all probation services in England and Wales. A similar tool is being embedded into courts and tribunal workflows with a pilot conducted in the Immigration and Asylum Chamber." }, { "id": "knife-crime-and-violence-model-kcvm", "title": "Knife Crime and Violence Model (KCVM)", "purpose": "Knife crime risk prediction", "url": "/ai-mapping/tools/knife-crime-and-violence-model-kcvm/", "deployment_stage": "Trialled", "development_type": "Academic collaboration", "developer_vendor": ["University of Essex"], "users": ["Essex"], "criminal_justice_stages": [1,2,3,7,8], "inference_modes": ["Synthesis"], "taxonomy": { "raw": ["Synthesis(enum^n -> enum)"], "parsed": [ { "original": "Synthesis(enum^n -> enum)", "inference_mode": "Synthesis", "inputs": [ "enum^n" ], "outputs": [ "enum" ] } ] }, "tool_functionality": ["future knife or violent crime risk profiling"], "resources": ["https://www.essex.police.uk/police-forces/essex-police/areas/essex-police/au/about-us/privacy-notices/knife-crime-and-violence-model--fearless-futures/","https://science.police.uk/delivery/case-studies/inside-essex-polices-battle-against-knife-violence/"], "notes": null }, { "id": "knife-crime-prediction-model", "title": "Knife crime prediction model", "purpose": "Knife crime location prediction", "url": "/ai-mapping/tools/knife-crime-prediction-model/", "deployment_stage": "Live", "development_type": "In-house", "developer_vendor": ["West Midlands"], "users": ["West Midlands"], "criminal_justice_stages": [1], "inference_modes": ["Synthesis"], "taxonomy": { "raw": ["Synthesis(enum^n -> enum)"], "parsed": [ { "original": "Synthesis(enum^n -> enum)", "inference_mode": "Synthesis", "inputs": [ "enum^n" ], "outputs": [ "enum" ] } ] }, "tool_functionality": ["crime hot spot"], "resources": ["https://foi.west-midlands.police.uk/wp-content/uploads/2022/08/855A_ATTACHMENT_01.pdf","https://www.westmidlands-pcc.gov.uk/wp-content/uploads/2023/06/2023_02_08-DAL-evaluation-of-4-week-predictive-tools-v1.0-1.pdf?x53331#page=6.12"], "notes": null }, { "id": "knife-crime-prediction", "title": "Knife Crime Prediction", "purpose": "Knife crime risk prediction", "url": "/ai-mapping/tools/knife-crime-prediction/", "deployment_stage": "Experimental", "development_type": "In-house", "developer_vendor": ["Thames Valley"], "users": ["Thames Valley"], "criminal_justice_stages": [1,2,7,8], "inference_modes": ["Synthesis"], "taxonomy": { "raw": ["Synthesis(enum^n -> enum)"], "parsed": [ { "original": "Synthesis(enum^n -> enum)", "inference_mode": "Synthesis", "inputs": [ "enum^n" ], "outputs": [ "enum" ] } ] }, "tool_functionality": ["future knife crime risk profiling"], "resources": ["https://science.police.uk/delivery/resources/predicting-knife-crime-an-individual-and-network-based-approach-thames-valley-police/"], "notes": null }, { "id": "knife-hunter", "title": "Knife Hunter", "purpose": "Knife type recognition", "url": "/ai-mapping/tools/knife-hunter/", "deployment_stage": "Live", "development_type": "Academic collaboration", "developer_vendor": ["University of Surrey"], "users": ["Metropolitan Police","Surrey","Lancashire","Northumbria","West Midlands","West Yorkshire","Greater Manchester","Suffolk","South Wales","Hampshire","Norfolk"], "criminal_justice_stages": [1,2,3], "inference_modes": ["Analysis"], "taxonomy": { "raw": ["Analysis(image -> enum)"], "parsed": [ { "original": "Analysis(image -> enum)", "inference_mode": "Analysis", "inputs": [ "image" ], "outputs": [ "enum" ] } ] }, "tool_functionality": ["knife type recognition"], "resources": ["https://www.surrey.ac.uk/news/artificial-intelligence-system-reshaping-uks-war-against-knife-crime"], "notes": "It has been deployed across 22 police forces. The forces being among the top monthly users are only noted." }, { "id": "knowledge-assistant-for-courts", "title": "Knowledge assistant for courts", "purpose": "Information searches for justice professionals", "url": "/ai-mapping/tools/knowledge-assistant-for-courts/", "deployment_stage": "Trialled", "development_type": "Unknown", "developer_vendor": [], "users": ["Courts"], "criminal_justice_stages": [5], "inference_modes": ["Generation"], "taxonomy": { "raw": ["Generation(prompt, enum -> text, [enum])"], "parsed": [ { "original": "Generation(prompt, enum -> text, [enum])", "inference_mode": "Generation", "inputs": [ "prompt" ,"enum" ], "outputs": [ "text" ,"[enum]" ] } ] }, "tool_functionality": ["generation of a text response, with a list of citations, given a prompt and named repository. Here, 1st enum is repository names and 2nd is citations (an example of RAG - retrieval augmented generation)"], "resources": ["https://ai.justice.gov.uk/our-work/semantic-search"], "notes": "Used by courts and tribunals staff." }, { "id": "linkage-analysis-tool-for-investigative-support-(latis)", "title": "Linkage Analysis Tool for Investigative Support (LATIS)", "purpose": "Crime linkage and investigative prioritisation", "url": "/ai-mapping/tools/linkage-analysis-tool-for-investigative-support-latis/", "deployment_stage": "Trialled", "development_type": "Academic collaboration", "developer_vendor": ["University of Leicester","University of Birmingham","Imperial College London","University of Brussels","National Crime Agency's Serious Crime Analysis Section (SCAS)"], "users": ["National Crime Agency's Serious Crime Analysis Section (SCAS)"], "criminal_justice_stages": [2,3], "inference_modes": ["Analysis"], "taxonomy": { "raw": ["Analysis([enum^n], [enum^n] -> enum)"], "parsed": [ { "original": "Analysis([enum^n], [enum^n] -> enum)", "inference_mode": "Analysis", "inputs": [ "[enum^n]" ,"[enum^n]" ], "outputs": [ "enum" ] } ] }, "tool_functionality": ["behavioural similarity of two offences represented by hundreds of behavioural variables"], "resources": ["https://link.springer.com/article/10.1007/s10940-025-09622-w","https://www.nationalcrimeagency.gov.uk/what-we-do/how-we-work/providing-specialist-capabilities-for-law-enforcement/serious-crime-analysis"], "notes": null }, { "id": "live-facial-recognition-lfr", "title": "Live Facial Recognition (LFR)", "purpose": "Locate persons of interest", "url": "/ai-mapping/tools/live-facial-recognition-lfr/", "deployment_stage": "Live", "development_type": "Third-party", "developer_vendor": ["Corsight AI","Digital Barriers","NEC"], "users": ["Essex","South Wales","Metropolitan Police","Northamptonshire","Greater Manchester","West Yorkshire","Bedfordshire","Surrey","Sussex","Thames Valley","Hampshire and Isle of Wight","Merseyside","Suffolk"], "criminal_justice_stages": [1,2,3], "inference_modes": ["Analysis"], "taxonomy": { "raw": ["Analysis(image, [image] -> enum, enum)"], "parsed": [ { "original": "Analysis(image, [image] -> enum, enum)", "inference_mode": "Analysis", "inputs": [ "image" ,"[image]" ], "outputs": [ "enum" ,"enum" ] } ] }, "tool_functionality": ["facial recognition (live) of unknown person appearing in video frame (image) based on a watchlist ([image]), where first enum is the index to the top match image from watchlist and second enum is the similarity score"], "resources": ["https://www.biometricupdate.com/202502/essex-police-reveal-impressive-accuracy-of-lfr-from-corsight-digital-barriers","https://www.essex.police.uk/police-forces/essex-police/areas/essex-police/au/about-us/live-facial-recognition/","https://www.biometricupdate.com/202502/essex-police-reveal-impressive-accuracy-of-lfr-from-corsight-digital-barriers","https://www.essex.police.uk/police-forces/essex-police/areas/essex-police/au/about-us/live-facial-recognition/","https://www.cardiff.ac.uk/__data/assets/pdf_file/0006/2426604/AFRReportDigital.pdf#page=6.21","https://www.south-wales.police.uk/police-forces/south-wales-police/areas/about-us/about-us/facial-recognition-technology/","https://www.met.police.uk/advice/advice-and-information/fr/facial-recognition-technology/","https://www.met.police.uk/SysSiteAssets/media/downloads/force-content/met/advice/lfr/policy-documents/lfr-policy-document2.pdf","https://www.bbc.co.uk/news/technology-69055945","https://science.police.uk/site/assets/files/3396/frt-equitability-study_mar2023.pdf","https://bigbrotherwatch.org.uk/press-releases/landmark-legal-challenges-launched-against-facial-recognition-after-police-and-retailer-misidentifications/","https://www.matrixlaw.co.uk/news/challenge-to-the-mets-use-of-live-facial-recognition-technology/","https://www.northants.police.uk/police-forces/northamptonshire-police/areas/northamptonshire-force-content/about-us/about-us/live-facial-recognition/","https://www.gov.uk/government/news/live-facial-recognition-technology-to-catch-high-harm-offenders","https://www.merseyside.police.uk/news/merseyside/news/2025/december-2025/police-roll-out-live-facial-recognition-in-merseyside-to-protect-communities/","https://emergencyservicestimes.com/2025/02/27/live-facial-recognition-trial-in-ipswich-secures-arrests/#:~:text=A%20trial%20of%20Live%20Facial,borrowed%20from%20neighbouring%20Essex%20Police."], "notes": null }, { "id": "magnet-automate", "title": "Magnet Automate", "purpose": "ID child sexual abuse/exploitation", "url": "/ai-mapping/tools/magnet-automate/", "deployment_stage": "Stage Unknown", "development_type": "Third-party", "developer_vendor": ["Magnet Forensics"], "users": ["Greater Manchester"], "criminal_justice_stages": [3], "inference_modes": ["Analysis"], "taxonomy": { "raw": ["Analysis(image -> enum)"], "parsed": [ { "original": "Analysis(image -> enum)", "inference_mode": "Analysis", "inputs": [ "image" ], "outputs": [ "enum" ] } ] }, "tool_functionality": ["identify images of child sexual abuse or of the sexual exploitation of children"], "resources": ["https://www.gmp.police.uk/news/greater-manchester/news/news/2023/august/gmps-award-winning-digital-forensic-investigation-unit-dfiu-are-breaking-new-ground-in-the-field-of-digital-forensics-relating-to-cse-and-child-sexual-abuse-images/","https://www.magnetforensics.com/wp-content/uploads/2022/11/MF_AUTOMATE_GMP_CaseStudy_8.5x11_Digital-1.pdf","https://www.facebook.com/MagnetForensics/videos/444076111194344/"], "notes": null }, { "id": "maltego-monitor", "title": "Maltego Monitor", "purpose": "Online sentiment analysis", "url": "/ai-mapping/tools/maltego-monitor/", "deployment_stage": "Stage Unknown", "development_type": "Third-party", "developer_vendor": ["Maltego"], "users": ["West Midlands","Greater Manchester","Metropolitan Police","South Wales"], "criminal_justice_stages": [1,2], "inference_modes": ["Analysis","Generation"], "taxonomy": { "raw": ["Analysis(text -> enum)","Generation(prompt, text -> text)"], "parsed": [ { "original": "Analysis(text -> enum)", "inference_mode": "Analysis", "inputs": [ "text" ], "outputs": [ "enum" ] }, { "original": "Generation(prompt, text -> text)", "inference_mode": "Generation", "inputs": [ "prompt" ,"text" ], "outputs": [ "text" ] } ] }, "tool_functionality": ["sentiment analysis","summarisation of data and provision of context from keyword-sentencing scripts"], "resources": ["https://www.maltego.com/law-enforcement/"], "notes": null }, { "id": "ndelius", "title": "nDelius", "purpose": "Information searches for probation practioners", "url": "/ai-mapping/tools/ndelius/", "deployment_stage": "Trialled", "development_type": "In-house", "developer_vendor": ["Ministry of Justice Data Science & AI Hub"], "users": ["Probation Service"], "criminal_justice_stages": [8], "inference_modes": ["Generation"], "taxonomy": { "raw": ["Generation(prompt, enum -> text, [enum])"], "parsed": [ { "original": "Generation(prompt, enum -> text, [enum])", "inference_mode": "Generation", "inputs": [ "prompt" ,"enum" ], "outputs": [ "text" ,"[enum]" ] } ] }, "tool_functionality": ["generation of a text response, with a list of citations, given a prompt and named repository. Here, 1st enum is repository names and 2nd is citations (an example of RAG - retrieval augmented generation)"], "resources": ["https://www.gov.uk/algorithmic-transparency-records/moj-ndelius-contact-log-semantic-search#name"], "notes": null }, { "id": "nectar", "title": "Nectar", "purpose": "Victim contact and update", "url": "/ai-mapping/tools/nectar/", "deployment_stage": "Trialled", "development_type": "Third-party", "developer_vendor": ["Palantir"], "users": ["Bedfordshire","Hertfordshire","Cambridgeshire","Derbyshire","Leicestershire","Lincolnshire","Northamptonshire","Nottinghamshire"], "criminal_justice_stages": [1,2,3], "inference_modes": ["Analysis","Synthesis"], "taxonomy": { "raw": ["Analysis(audio -> text)","Synthesis(enum^n -> enum)"], "parsed": [ { "original": "Analysis(audio -> text)", "inference_mode": "Analysis", "inputs": [ "audio" ], "outputs": [ "text" ] }, { "original": "Synthesis(enum^n -> enum)", "inference_mode": "Synthesis", "inputs": [ "enum^n" ], "outputs": [ "enum" ] } ] }, "tool_functionality": ["transcription","THRIVE (threat, harm risk, investigation and engagement) risk profiling for emergency call"], "resources": ["https://medium.com/@junghoonchoi_20153/transforming-law-enforcement-how-ai-is-changing-the-fight-against-crime-919bf3c16c2e","https://www.whatdotheyknow.com/request/use_of_palantir_technologies_sof_2/response/2196967/attach/html/3/BCH%20Response%20Letter%20FOI2022%2007166%2007164%2007160.docx.html","https://www.rocu.police.uk/news/2025/july/police-pilot-advanced-data-platform-to-tackle-serious-and-organised-crime/","https://libertyinvestigates.org.uk/articles/uk-police-working-with-controversial-tech-giant-palantir-on-real-time-surveillance-network/","https://www.rocu.police.uk/news/2025/july/police-pilot-advanced-data-platform-to-tackle-serious-and-organised-crime/"], "notes": null }, { "id": "nice", "title": "NiCE", "purpose": "Digital evidence analysis", "url": "/ai-mapping/tools/nice/", "deployment_stage": "Stage Unknown", "development_type": "Third-party", "developer_vendor": ["NiCE"], "users": ["West Midlands"], "criminal_justice_stages": [1,2,3], "inference_modes": ["Analysis"], "taxonomy": { "raw": ["Analysis(video -> [enum])","Analysis(audio -> text)","Analysis(image -> text)","Analysis(text -> text)","Analysis([video], [image], [text], [audio] -> [enum])","Analysis([video], [image], [text], [audio] -> [enum, enum])"], "parsed": [ { "original": "Analysis(video -> [enum])", "inference_mode": "Analysis", "inputs": [ "video" ], "outputs": [ "[enum]" ] }, { "original": "Analysis(audio -> text)", "inference_mode": "Analysis", "inputs": [ "audio" ], "outputs": [ "text" ] }, { "original": "Analysis(image -> text)", "inference_mode": "Analysis", "inputs": [ "image" ], "outputs": [ "text" ] }, { "original": "Analysis(text -> text)", "inference_mode": "Analysis", "inputs": [ "text" ], "outputs": [ "text" ] }, { "original": "Analysis([video], [image], [text], [audio] -> [enum])", "inference_mode": "Analysis", "inputs": [ "[video]" ,"[image]" ,"[text]" ,"[audio]" ], "outputs": [ "[enum]" ] }, { "original": "Analysis([video], [image], [text], [audio] -> [enum, enum])", "inference_mode": "Analysis", "inputs": [ "[video]" ,"[image]" ,"[text]" ,"[audio]" ], "outputs": [ "[enum" ,"enum]" ] } ] }, "tool_functionality": ["face detection","audio transcription","object character recognition (OCR)","translation","query all indexed data based on incident parameters and recommend additional evidence","query all indexed data based on incident parameters to uncover hidden connections"], "resources": ["https://www.nicepublicsafety.com/nice-ai"], "notes": null }, { "id": "nlp-to-detect-threatening-and-abusive-language-toward-victims", "title": "NLP to detect threatening and abusive language toward victims", "purpose": "Online sexism detection", "url": "/ai-mapping/tools/nlp-to-detect-threatening-and-abusive-language-toward-victims/", "deployment_stage": "Experimental", "development_type": "Academic collaboration", "developer_vendor": ["Warwick University"], "users": ["Forensic Capability Network (FCN)"], "criminal_justice_stages": [1,2,3], "inference_modes": ["Analysis"], "taxonomy": { "raw": ["Analysis(text -> [enum])"], "parsed": [ { "original": "Analysis(text -> [enum])", "inference_mode": "Analysis", "inputs": [ "text" ], "outputs": [ "[enum]" ] } ] }, "tool_functionality": ["feature extraction in NLP"], "resources": ["https://science.police.uk/opportunities/police-star-fund/previously-funded-police-star-fund-projects/","https://www.fcn.police.uk/news/2023-09/police-forensic-experts-trial-ai-detect-online-sexism"], "notes": null }, { "id": "offender-escalation-model-stalking-and-harassment", "title": "Offender Escalation Model (Stalking and Harassment)", "purpose": "Offender risk prediction", "url": "/ai-mapping/tools/offender-escalation-model-stalking-and-harassment/", "deployment_stage": "Stage Unknown", "development_type": "In-house", "developer_vendor": ["West Midlands"], "users": ["West Midlands"], "criminal_justice_stages": [1,2,3,7,8], "inference_modes": ["Synthesis"], "taxonomy": { "raw": ["Synthesis(enum^n -> enum)"], "parsed": [ { "original": "Synthesis(enum^n -> enum)", "inference_mode": "Synthesis", "inputs": [ "enum^n" ], "outputs": [ "enum" ] } ] }, "tool_functionality": ["classify offender based on their crime history"], "resources": ["https://www.westmidlands-pcc.gov.uk/wp-content/uploads/2023/08/2023_09-main_report_Stalker_Risk_FINAL.pdf?x54376"], "notes": null }, { "id": "offender-time-to-event-model-stalking-and-harassment", "title": "Offender Time-to-Event Model (Stalking and Harassment)", "purpose": "Time-based offender risk prediction", "url": "/ai-mapping/tools/offender-time-to-event-model-stalking-and-harassment/", "deployment_stage": "Stage Unknown", "development_type": "In-house", "developer_vendor": ["West Midlands"], "users": ["West Midlands"], "criminal_justice_stages": [1,2,3,7,8], "inference_modes": ["Synthesis"], "taxonomy": { "raw": ["Synthesis(enum^n -> enum)"], "parsed": [ { "original": "Synthesis(enum^n -> enum)", "inference_mode": "Synthesis", "inputs": [ "enum^n" ], "outputs": [ "enum" ] } ] }, "tool_functionality": ["prediction of time-to-event (survival analysis)"], "resources": ["https://www.westmidlands-pcc.gov.uk/wp-content/uploads/2023/08/2023_09-main_report_Stalker_Risk_FINAL.pdf?x54378"], "notes": null }, { "id": "on-premises-gen-ai-solution", "title": "On-premises Gen-AI solution", "purpose": "AI-assisted policing", "url": "/ai-mapping/tools/on-premises-gen-ai-solution/", "deployment_stage": "Trialled", "development_type": "In-house", "developer_vendor": ["Surrey","Sussex"], "users": ["Surrey","Sussex"], "criminal_justice_stages": [3], "inference_modes": ["Analysis","Generation"], "taxonomy": { "raw": ["Analysis(audio -> text)","Generation(prompt, text -> text)"], "parsed": [ { "original": "Analysis(audio -> text)", "inference_mode": "Analysis", "inputs": [ "audio" ], "outputs": [ "text" ] }, { "original": "Generation(prompt, text -> text)", "inference_mode": "Generation", "inputs": [ "prompt" ,"text" ], "outputs": [ "text" ] } ] }, "tool_functionality": ["transcription","summarisation, redaction"], "resources": [], "notes": null }, { "id": "opentext-ediscovery-axcelerate", "title": "OpenText eDiscovery (Axcelerate)", "purpose": "AI-assisted document review", "url": "/ai-mapping/tools/opentext-ediscovery-axcelerate/", "deployment_stage": "Trialled", "development_type": "Third-party", "developer_vendor": ["OpenText"], "users": ["Serious Fraud Office"], "criminal_justice_stages": [3,4,5], "inference_modes": ["Generation"], "taxonomy": { "raw": ["Generation(prompt, text -> text)","Generation(prompt, text -> [text, enum])","Generation(prompt, text, [enum] -> text)","Generation(prompt, [text], [enum] -> [enum, enum])"], "parsed": [ { "original": "Generation(prompt, text -> text)", "inference_mode": "Generation", "inputs": [ "prompt" ,"text" ], "outputs": [ "text" ] }, { "original": "Generation(prompt, text -> [text, enum])", "inference_mode": "Generation", "inputs": [ "prompt" ,"text" ], "outputs": [ "[text" ,"enum]" ] }, { "original": "Generation(prompt, text, [enum] -> text)", "inference_mode": "Generation", "inputs": [ "prompt" ,"text" ,"[enum]" ], "outputs": [ "text" ] }, { "original": "Generation(prompt, [text], [enum] -> [enum, enum])", "inference_mode": "Generation", "inputs": [ "prompt" ,"[text]" ,"[enum]" ], "outputs": [ "[enum" ,"enum]" ] } ] }, "tool_functionality": ["summarisation","retrieval of key information from documents","redaction of document given a list of information to be redacted","analysis of a collection of digital documents as to the degree of their responsiveness to a set of disclosure tests"], "resources": ["https://assets.publishing.service.gov.uk/media/67e279b64fed20c7f559f558/Disclosure_in_the_Digital_Age_-_Independent_Review_Report_CP1285_WEB_0.1.pdf","https://www.parallelparliament.co.uk/question/26405/serious-fraud-office-artificial-intelligence#:~:text=Answered%20by&text=Artificial%20Intelligence%20(AI)%20is%20at,SFO%20cases%20in%20the%20future","https://blogs.opentext.com/whats-new-in-opentext-ediscovery/","https://www.opentext.com/what-is/document-review","https://www.opentext.com/products/ediscovery"], "notes": null }, { "id": "operator-initiated-facial-recognition-oifr", "title": "Operator Initiated Facial Recognition (OIFR)", "purpose": "ID people at risk", "url": "/ai-mapping/tools/operator-initiated-facial-recognition-oifr/", "deployment_stage": "Live", "development_type": "Third-party", "developer_vendor": ["NEC"], "users": ["South Wales","Gwent","Thames Valley","Hampshire and Isle of Wight"], "criminal_justice_stages": [1,2,3], "inference_modes": ["Analysis"], "taxonomy": { "raw": ["Analysis(image, [image] -> [enum, enum])"], "parsed": [ { "original": "Analysis(image, [image] -> [enum, enum])", "inference_mode": "Analysis", "inputs": [ "image" ,"[image]" ], "outputs": [ "[enum" ,"enum]" ] } ] }, "tool_functionality": ["facial recognition (operator initiated) of unknown person appearing in still image based on a reference image database ([image]), where first enum is the index to an image in the reference database and second enum is the similarity score"], "resources": ["https://www.south-wales.police.uk/news/south-wales/news/2024/december/welsh-police-forces-launch-first-facial-recognition-mobile-app/","https://www.south-wales.police.uk/police-forces/south-wales-police/areas/about-us/about-us/facial-recognition-technology/faqs-for-operator-initiated-facial-recognition-app/","https://science.police.uk/site/assets/files/3396/frt-equitability-study_mar2023.pdf","https://www.gov.uk/government/publications/police-use-of-facial-recognition/police-use-of-facial-recognition-factsheet"], "notes": null }, { "id": "patrol-wise", "title": "Patrol-Wise", "purpose": "Street-level burglary prediction", "url": "/ai-mapping/tools/patrol-wise/", "deployment_stage": "Live", "development_type": "Academic collaboration", "developer_vendor": ["University College London (UCL)"], "users": ["West Yorkshire"], "criminal_justice_stages": [1,2], "inference_modes": ["Synthesis"], "taxonomy": { "raw": ["Synthesis(enum^n -> enum)"], "parsed": [ { "original": "Synthesis(enum^n -> enum)", "inference_mode": "Synthesis", "inputs": [ "enum^n" ], "outputs": [ "enum" ] } ] }, "tool_functionality": ["crime hot spot"], "resources": [], "notes": null }, { "id": "profile-unmasking-and-linking-software-engine-pulse", "title": "Profile Unmasking and Linking Software Engine (PULSE)", "purpose": "Online identity verification and crime detection", "url": "/ai-mapping/tools/profile-unmasking-and-linking-software-engine-pulse/", "deployment_stage": "Stage Unknown", "development_type": "Academic collaboration", "developer_vendor": ["University of Hertfordshire"], "users": ["Hertfordshire"], "criminal_justice_stages": [1,2,3], "inference_modes": ["Analysis"], "taxonomy": { "raw": ["Analysis(text -> enum)"], "parsed": [ { "original": "Analysis(text -> enum)", "inference_mode": "Analysis", "inputs": [ "text" ], "outputs": [ "enum" ] } ] }, "tool_functionality": ["sentiment analysis, content analysis"], "resources": ["https://science.police.uk/site/assets/files/3795/pulse_profile_unmasking_linking_software_engine_-_complete.pdf"], "notes": null }, { "id": "qlik-sense", "title": "Qlik Sense", "purpose": "Offender/victim risk prediction", "url": "/ai-mapping/tools/qlik-sense/", "deployment_stage": "Live", "development_type": "Third-party", "developer_vendor": ["Qlik"], "users": ["Avon and Somerset"], "criminal_justice_stages": [1,2,7,8], "inference_modes": ["Synthesis"], "taxonomy": { "raw": ["Synthesis(enum^n -> enum)"], "parsed": [ { "original": "Synthesis(enum^n -> enum)", "inference_mode": "Synthesis", "inputs": [ "enum^n" ], "outputs": [ "enum" ] } ] }, "tool_functionality": ["offence risk profiling"], "resources": ["https://www.avonandsomerset.police.uk/about/policies-and-procedures/data-science/","https://www.qlik.com/us/products/qlik-sense"], "notes": null }, { "id": "rapid-video-response-rvr-with-ai-overlay", "title": "Rapid Video Response (RVR) with AI Overlay", "purpose": "Witness statement generation from domestic response", "url": "/ai-mapping/tools/rapid-video-response-rvr-with-ai-overlay/", "deployment_stage": "Trialled", "development_type": "Unknown", "developer_vendor": [], "users": ["Bedfordshire","Cambridgeshire","Essex","Hertfordshire","Kent","Suffolk","Norfolk"], "criminal_justice_stages": [1,2,3], "inference_modes": ["Analysis","Generation"], "taxonomy": { "raw": ["Analysis(audio -> text)","Generation(prompt, text -> text)"], "parsed": [ { "original": "Analysis(audio -> text)", "inference_mode": "Analysis", "inputs": [ "audio" ], "outputs": [ "text" ] }, { "original": "Generation(prompt, text -> text)", "inference_mode": "Generation", "inputs": [ "prompt" ,"text" ], "outputs": [ "text" ] } ] }, "tool_functionality": ["transcription","witness statements writing"], "resources": ["https://www.college.police.uk/support-forces/practices/rapid-video-response-rvr-domestic-abuse"], "notes": null }, { "id": "retrospective-facial-recognition-rfr", "title": "Retrospective Facial Recognition (RFR)", "purpose": "ID offenders", "url": "/ai-mapping/tools/retrospective-facial-recognition-rfr/", "deployment_stage": "Live", "development_type": "Third-party", "developer_vendor": ["NEC"], "users": ["West Mercia","West Midlands","South Wales","Metropolitan Police","Thames Valley","Hampshire and Isle of Wight"], "criminal_justice_stages": [2,3], "inference_modes": ["Analysis"], "taxonomy": { "raw": ["Analysis(image, [image] -> [enum, enum])"], "parsed": [ { "original": "Analysis(image, [image] -> [enum, enum])", "inference_mode": "Analysis", "inputs": [ "image" ,"[image]" ], "outputs": [ "[enum" ,"enum]" ] } ] }, "tool_functionality": ["facial recognition (retrospective) of unknown person appearing in still image based on a reference image database ([image]), where first enum is the index to an image in the reference database and second enum is the similarity score"], "resources": ["https://www.westmercia.police.uk/foi-ai/west-mercia-police/2021/december/foi-141620/","https://museum.west-midlands.police.uk/general/technology-facial-recognition/","https://www.westmercia.police.uk/foi-ai/west-mercia-police/2021/december/foi-141620/","https://museum.west-midlands.police.uk/general/technology-facial-recognition/","https://www.cardiff.ac.uk/__data/assets/pdf_file/0006/2426604/AFRReportDigital.pdf#page=6.21","https://www.south-wales.police.uk/police-forces/south-wales-police/areas/about-us/about-us/facial-recognition-technology/","https://www.gov.uk/government/publications/police-use-of-facial-recognition/police-use-of-facial-recognition-factsheet"], "notes": null }, { "id": "risk-terrain-modelling-rtm-rtmdx", "title": "Risk Terrain Modelling (RTM) - RTMDx", "purpose": "Geospatial crime prediction", "url": "/ai-mapping/tools/risk-terrain-modelling-rtm-rtmdx/", "deployment_stage": "Live", "development_type": "Third-party", "developer_vendor": ["Unknown"], "users": ["Essex","Metropolitan Police","Merseyside"], "criminal_justice_stages": [1,2], "inference_modes": ["Synthesis"], "taxonomy": { "raw": ["Synthesis(enum^n -> enum)"], "parsed": [ { "original": "Synthesis(enum^n -> enum)", "inference_mode": "Synthesis", "inputs": [ "enum^n" ], "outputs": [ "enum" ] } ] }, "tool_functionality": ["crime hot spot"], "resources": ["https://www.rutgerscps.org/uploads/2/7/3/7/27370595/essexuk_rbp2023_infographic.pdf","https://www.met.police.uk/foi-ai/metropolitan-police/disclosure-2023/june-2023/mps-use-risk-terrain-modelling-application/","https://www.merseyside.police.uk/news/merseyside/news/2023/september/merseyside-police-adopts-geospatial-technology-to-support-crime-prevention/","https://www.amnesty.org.uk/files/2025-02/Automated%20Racism%20Report%20-%20Amnesty%20International%20UK%20-%202025.pdf?VersionId=JqCcTODw37yAXyINmAY6uAzrKEWucFF7#page=80.05"], "notes": null }, { "id": "riven-docdefender", "title": "Riven DocDefender", "purpose": "Document redaction", "url": "/ai-mapping/tools/riven-docdefender/", "deployment_stage": "Live", "development_type": "Third-party", "developer_vendor": ["Riven"], "users": ["Bedfordshire","Hertfordshire","Cambridgeshire","Kent","Essex","Hampshire and Isle of Wight","Thames Valley"], "criminal_justice_stages": [3], "inference_modes": ["Analysis"], "taxonomy": { "raw": ["Analysis(text -> text)"], "parsed": [ { "original": "Analysis(text -> text)", "inference_mode": "Analysis", "inputs": [ "text" ], "outputs": [ "text" ] } ] }, "tool_functionality": ["text redaction"], "resources": ["https://www.beds.police.uk/news/bedfordshire/news/2023/12-december/cutting-edge-tech-saving-bedfordshire-police-officers-time/#:~:text=Cutting%20edge%20technology%20has%20been,shifts%20to%20a%20few%20minutes."], "notes": null }, { "id": "secureredact", "title": "SecureRedact", "purpose": "People, object and text redaction in videos", "url": "/ai-mapping/tools/secureredact/", "deployment_stage": "Trialled", "development_type": "Third-party", "developer_vendor": ["Pimloc"], "users": ["Sussex","Avon and Somerset","Humberside","Metropolitan Police"], "criminal_justice_stages": [3], "inference_modes": ["Analysis"], "taxonomy": { "raw": ["Analysis(audio -> text)","Analysis(text -> text)","Analysis(video, image -> [image, enum])","Analysis(video -> video)","Analysis(audio, [audio, enum] -> [audio, enum])","Analysis(audio -> audio)"], "parsed": [ { "original": "Analysis(audio -> text)", "inference_mode": "Analysis", "inputs": [ "audio" ], "outputs": [ "text" ] }, { "original": "Analysis(text -> text)", "inference_mode": "Analysis", "inputs": [ "text" ], "outputs": [ "text" ] }, { "original": "Analysis(video, image -> [image, enum])", "inference_mode": "Analysis", "inputs": [ "video" ,"image" ], "outputs": [ "[image" ,"enum]" ] }, { "original": "Analysis(video -> video)", "inference_mode": "Analysis", "inputs": [ "video" ], "outputs": [ "video" ] }, { "original": "Analysis(audio, [audio, enum] -> [audio, enum])", "inference_mode": "Analysis", "inputs": [ "audio" ,"[audio" ,"enum]" ], "outputs": [ "[audio" ,"enum]" ] }, { "original": "Analysis(audio -> audio)", "inference_mode": "Analysis", "inputs": [ "audio" ], "outputs": [ "audio" ] } ] }, "tool_functionality": ["transcription","translation","face/object recognition","face/object/text redaction in videos","speaker diarisation","audio redaction"], "resources": ["https://www.secureredact.ai/"], "notes": null }, { "id": "soze-ai", "title": "Soze AI", "purpose": "Cold case analysis", "url": "/ai-mapping/tools/soze-ai/", "deployment_stage": "Trialled", "development_type": "Third-party", "developer_vendor": ["Akkodis"], "users": ["Avon and Somerset"], "criminal_justice_stages": [1,2,3], "inference_modes": ["Analysis"], "taxonomy": { "raw": ["Analysis(text, [enum] -> [enum, enum])","Analysis(video, image -> [enum])"], "parsed": [ { "original": "Analysis(text, [enum] -> [enum, enum])", "inference_mode": "Analysis", "inputs": [ "text" ,"[enum]" ], "outputs": [ "[enum" ,"enum]" ] }, { "original": "Analysis(video, image -> [enum])", "inference_mode": "Analysis", "inputs": [ "video" ,"image" ], "outputs": [ "[enum]" ] } ] }, "tool_functionality": ["search for keywords in text, where first enum is the keyword list, second enum is a keyword from the list, and third enum is the index (i.e. position) in text","find person or object in video"], "resources": ["https://news.sky.com/story/ai-tool-that-can-do-81-years-of-detective-work-in-30-hours-trialled-by-police-13220891","https://www.akkodis.com/en/tech-practices/ai-solutions-data-analytics/soze-solution-platform"], "notes": null }, { "id": "stalking-behaviour-detection-tool", "title": "Stalking behaviour detection tool", "purpose": "Stalking behaviours identification", "url": "/ai-mapping/tools/stalking-behaviour-detection-tool/", "deployment_stage": "Experimental", "development_type": "In-house", "developer_vendor": ["Cheshire"], "users": ["Cheshire"], "criminal_justice_stages": [1,2,3], "inference_modes": ["Analysis"], "taxonomy": { "raw": ["Analysis(text -> enum)"], "parsed": [ { "original": "Analysis(text -> enum)", "inference_mode": "Analysis", "inputs": [ "text" ], "outputs": [ "enum" ] } ] }, "tool_functionality": ["identify stalking behaviours"], "resources": ["https://www.cheshire-pcc.gov.uk/news/latest-news/2025/05/cheshire-leads-the-way-with-technology-to-help-identify-stalking-behaviours-sooner/"], "notes": null }, { "id": "toex-translate-and-transcribe-tool", "title": "TOEX Translate and Transcribe tool", "purpose": "Translation & transcription", "url": "/ai-mapping/tools/toex-translate-and-transcribe-tool/", "deployment_stage": "Trialled", "development_type": "Third-party", "developer_vendor": ["TOEX","Microsoft"], "users": ["Norfolk","Suffolk"], "criminal_justice_stages": [4], "inference_modes": ["Analysis"], "taxonomy": { "raw": ["Analysis(audio -> text)","Analysis(text, text)","Analysis(text, enum -> enum)","Analysis(text, [enum] -> [enum, enum])"], "parsed": [ { "original": "Analysis(audio -> text)", "inference_mode": "Analysis", "inputs": [ "audio" ], "outputs": [ "text" ] }, { "original": "Analysis(text, text)", "inference_mode": "Analysis", "inputs": [ "text" ,"text" ], "outputs": [ ] }, { "original": "Analysis(text, enum -> enum)", "inference_mode": "Analysis", "inputs": [ "text" ,"enum" ], "outputs": [ "enum" ] }, { "original": "Analysis(text, [enum] -> [enum, enum])", "inference_mode": "Analysis", "inputs": [ "text" ,"[enum]" ], "outputs": [ "[enum" ,"enum]" ] } ] }, "tool_functionality": ["transcription","language translation","named entity recognition","search for keywords in text, where first enum is the keyword list, second enum is a keyword from the list, and third enum is the index (i.e. position) in text"], "resources": ["https://www.toexprogramme.co.uk/assets/TOEX-E-Magazine-9-April-June-2024.pdf","https://science.police.uk/site/assets/files/4619/toex_e-magazine_6_july_-_september_2023.pdf","https://www.toexprogramme.co.uk/assets/TOEX-Infographic-Tech-Enablers-FINAL.pdf"], "notes": null }, { "id": "tuserv", "title": "tuServ", "purpose": "Statements writing through narration", "url": "/ai-mapping/tools/tuserv/", "deployment_stage": "Trialled", "development_type": "Third-party", "developer_vendor": ["Black Marble"], "users": ["Hertfordshire","Cambridgeshire"], "criminal_justice_stages": [3], "inference_modes": ["Analysis"], "taxonomy": { "raw": ["Analysis(audio -> text)"], "parsed": [ { "original": "Analysis(audio -> text)", "inference_mode": "Analysis", "inputs": [ "audio" ], "outputs": [ "text" ] } ] }, "tool_functionality": ["transcription"], "resources": ["https://tuserv.com/","https://tuserv.com/case-studies-and-media/catching-criminals-with-uwp/"], "notes": null }, { "id": "untrite-thrive-ai", "title": "Untrite Thrive AI", "purpose": "AI-assisted emergency call handling", "url": "/ai-mapping/tools/untrite-thrive-ai/", "deployment_stage": "Trialled", "development_type": "Third-party", "developer_vendor": ["Untrite"], "users": ["Humberside"], "criminal_justice_stages": [1,2,3], "inference_modes": ["Analysis","Generation","Synthesis"], "taxonomy": { "raw": ["Analysis(audio -> text)","Generation(prompt, text, [enum] -> [text])","Synthesis(enum^n -> enum)","Generation(prompt, enum -> text)"], "parsed": [ { "original": "Analysis(audio -> text)", "inference_mode": "Analysis", "inputs": [ "audio" ], "outputs": [ "text" ] }, { "original": "Generation(prompt, text, [enum] -> [text])", "inference_mode": "Generation", "inputs": [ "prompt" ,"text" ,"[enum]" ], "outputs": [ "[text]" ] }, { "original": "Synthesis(enum^n -> enum)", "inference_mode": "Synthesis", "inputs": [ "enum^n" ], "outputs": [ "enum" ] }, { "original": "Generation(prompt, enum -> text)", "inference_mode": "Generation", "inputs": [ "prompt" ,"enum" ], "outputs": [ "text" ] } ] }, "tool_functionality": ["transcription","search for keywords and return list of text entities that could be search inputs to the police database, where enum is the categories of entities that can be searched, e.g. name, address","THRIVE (threat, harm risk, investigation and engagement) risk profiling for emergency call","draft an assessment using the data of the threat, harm, risk and vulnerability of that particular incident that’s being reported"], "resources": ["https://pds.police.uk/press-release-artificial-intelligence-introduced-to-enhance-emergency-response/","https://emergencyservicestimes.com/2024/01/10/ai-as-a-personal-assistant-in-the-control-room/","https://www.college.police.uk/guidance/vulnerability-related-risks/introduction-vulnerability-related-risk#thrive","https://news.npcc.police.uk/editorial/ai-in-policing-innovating-public-centric-digital-services","https://www.ukauthority.com/articles/humberside-police-tests-ai-for-domestic-abuse-calls/","https://untrite.com/case-study/"], "notes": null }, { "id": "victim-harm-model-stalking-and-harassment", "title": "Victim Harm Model (Stalking and Harassment)", "purpose": "Victim risk prediction", "url": "/ai-mapping/tools/victim-harm-model-stalking-and-harassment/", "deployment_stage": "Stage Unknown", "development_type": "In-house", "developer_vendor": ["West Midlands"], "users": ["West Midlands"], "criminal_justice_stages": [1], "inference_modes": ["Synthesis"], "taxonomy": { "raw": ["Synthesis(enum^n -> enum)"], "parsed": [ { "original": "Synthesis(enum^n -> enum)", "inference_mode": "Synthesis", "inputs": [ "enum^n" ], "outputs": [ "enum" ] } ] }, "tool_functionality": ["classify victim based on their previous victim experiences"], "resources": ["https://www.westmidlands-pcc.gov.uk/wp-content/uploads/2023/08/2023_09-main_report_Stalker_Risk_FINAL.pdf?x54377"], "notes": null }, { "id": "victim-time-to-event-model-stalking-and-harassment", "title": "Victim Time-to-Event Model (Stalking and Harassment)", "purpose": "Time-based victim risk prediction", "url": "/ai-mapping/tools/victim-time-to-event-model-stalking-and-harassment/", "deployment_stage": "Stage Unknown", "development_type": "In-house", "developer_vendor": ["West Midlands"], "users": ["West Midlands"], "criminal_justice_stages": [1], "inference_modes": ["Synthesis"], "taxonomy": { "raw": ["Synthesis(enum^n -> enum)"], "parsed": [ { "original": "Synthesis(enum^n -> enum)", "inference_mode": "Synthesis", "inputs": [ "enum^n" ], "outputs": [ "enum" ] } ] }, "tool_functionality": ["prediction of time-to-event (survival analysis)"], "resources": ["https://www.westmidlands-pcc.gov.uk/wp-content/uploads/2023/08/2023_09-main_report_Stalker_Risk_FINAL.pdf?x54379"], "notes": null }, { "id": "vigil-ai", "title": "Vigil AI", "purpose": "Vulnerable people protection", "url": "/ai-mapping/tools/vigil-ai/", "deployment_stage": "Live", "development_type": "Third-party", "developer_vendor": ["ROKE"], "users": [], "criminal_justice_stages": [1,2,3], "inference_modes": ["Analysis"], "taxonomy": { "raw": ["Analysis(image -> enum)"], "parsed": [ { "original": "Analysis(image -> enum)", "inference_mode": "Analysis", "inputs": [ "image" ], "outputs": [ "enum" ] } ] }, "tool_functionality": ["detecting and categorising Child Sexual Abuse Material (CSAM) including both images and videos"], "resources": ["https://www.roke.co.uk/products/vigil-ai"], "notes": null } ] }