AIPAS plans a series of outputs to increase the visibility and impact of the project in the policing and security space.
Accountability Framework
AIPAS will contextualise the AP4AI Accountability Framework for the UK policing and security ecosystem. The framework will be based on comprehensive, domain-expert and citizen validations and feed into UK Government efforts on Responsible AI.
The public report will be available later in the project.
Implementation Guidelines
Technology-agnostic implementation guidelines for AI Accountability in the UK policing and security domain including the steps, stages and processes policing and security organisations need to take to demonstrate AI Accountability in line with the UK Government AI Strategy and societal expectations.
The implementation guidelines will be available to UK policing and security organisations only.
Use Case Guidelines
AI Accountability guidance for four use cases in core AI application areas for LEAs:
- forensics,
- counter-terrorism,
- fight against child sexual exploitation, and
- serious and organised crime.
The use cases will address and explore the notion of AI Accountability for the extended policing and security value chain to include providers, suppliers and developers of AI services and capabilities.
The implementation guidelines will be available to UK policing and security organisations only.
Citizen Consultation
A comprehensive citizen consultation will lead to insights into expectations about mechanisms for AI Accountability implementation in the policing and security ecosystem.
The public report will be available later in the project.
Software Tool
A freely available, robust and secure software tool based on UK government ICT Standard and deployable on government cloud.
The software tool will be developed to TRL 8.
Public Awareness and Outreach
AIPAS will implement an online outreach programme for public awareness and education on AI accountability.
Publications
Find here our publications emerging from the project.
AP4AI
The Accountability Principles for AI (AP4AI) project develops solutions to assess, review and safeguard the accountability of AI usage by internal security practitioners in line with EU values and fundamental rights.
Find here reports published by the AP4AI project.
International Citizen Consultation on AI Accountability in Policing
Read the report: International Citizen Consultation on AI Accountability in Policing
AP4AI’s International Citizen Consultation was conducted in 30 countries, including all 27 EU Member States, the USA, UK and Australia. The consultation report offers detailed descriptions of public reactions to AI use by police as well as the concrete mechanisms citizens request and expect to ensure AI Accountability.
AP4AI Framework Blueprint
Read the report: AP4AI Framework Blueprint
This report presents the Blueprint of the AP4AI Framework, with a special focus on proposals for application methods. It further includes a narrative review of current discourse on AI frameworks and regulations relevant for AP4AI, as well as high-level findings of an ongoing citizen consultation across 30 countries.
Summary Report on Expert Consultations
Read the report: Summary Report on Expert Consultations
This report is intended as a reference document to establish the foundation of Accountability for AI in the Internal Security Domain. It defines the initial set of 12 AP4AI Principles developed by the AP4AI consortium in consultations with international AI experts in the first phase of the AP4AI project.