How the CPS will use Artificial Intelligence

We will maintain a 'Human in the Loop'

We will ensure adequate human participation in AI processes and verify the accuracy of outputs by an appropriately qualified and trained person. Individuals are responsible for each use and output generated using AI, just as they would be if an AI tool had not been used. Individual tools will have a Senior Responsible Officer and a person or individual will be responsible for each use. We will support and empower our staff in the use of AI and provide training to ensure staff are confident with the organisation’s use of AI, the use of AI within their roles and that they understand the systems, security and processes associated with AI.

We will also put mechanisms in place for end-users to provide feedback on system performance and accuracy.

We will ensure that our use of AI is communicated clearly to our staff, police, key Criminal Justice System and legal partners, victims, suspects, and the general public.

By being transparent about the tools we use and how our data and evidence are utilised, we aim to increase public confidence, protect data, and mitigate reputational risks.

At each stage of AI’s use, from proof of concept to post implementation, evaluation will take place.

We will monitor and assess these processes to identify any unintended risk and harms associated with tools at all phases and this will ensure that we can continue to use the AI for its intended purpose whilst mitigating against risks and harms. Feedback will be taken by users at all stages and we will address risks, concerns or harms as they arise.

We will collaborate with partners on cross-government aims and wider priorities, including the Government’s AI Strategy and Playbook.

We will work with our partner agencies to understand where their use of AI may impact legal outcomes and processes and will collaborate across government and law enforcement to facilitate end-to-end alignment across the CJS and establish opportunities for sharing knowledge and best practices. This includes working with CJS partners to ensure our vision and broader use of AI are aligned with key considerations, for example the Policing AI Covenant.

We will have robust processes setting out how we are using AI, why we are using it and the outcomes associated with each AI tool.

Artificial Intelligence cases are currently evaluated through a framework which starts with a problem and explores all potential solutions. This ensures that AI offers genuine business solutions and is deployed efficiently. AI should always be used to enhance efficiency, increase quality and deliver a good service.

To ensure this is the case and that our use of AI is reliable and secure, organisationally, we must be able to articulate the functionality of our tools. This will allow us and our people to observe the approaches set out in this vision, using AI responsibly, transparently, securely and efficiently.

AI will be used responsibly by the CPS, in accordance with the AI Data Security and Ethics Principles alongside other internal checks and balances.

The key principles of CPS’s AI Data Security and Ethics Framework emphasise understanding AI's capabilities and risks, ensuring AI projects are driven by user needs and public benefits, and using AI in a lawful, fair, and ethical manner. Transparency, accountability, and human oversight are crucial, along with ensuring AI security and strong governance. Collaboration, skills development, and considering the environmental impact of AI are also important. Finally, managing the full lifecycle of AI, including maintenance and updates, is essential.