The Department for Science, Innovation and Technology (DSIT) has published a new AI Playbook on building AI into public services.
It includes a series of tips and 10 principles for public servants to understand what AI can and cannot do and how to mitigate the risks involved.
Technology Secretary Peter Kyle said: “The publication of our AI Playbook today comes with a call to arms for tech specialists across the public sector – use the guidance we are sharing to put AI to work in your organisations at whiplash speed, so we can repair our broken public services together.”
The document describes AI as comprising an evolving set of interconnected fields including machine learning, deep learning, language models, foundation models and generative AI, and emphasises the importance of keeping sight of the ethics and societal impacts.
This provides the basis for its 10 principles:
- Know what AI is and what its limitations are.
- Use AI lawfully, ethically and responsibly.
- Know how to use AI securely.
- Have meaningful human control at the right stage.
- Understand how to manage the AI life cycle.
- Use the right tool for the job.
- Be open and collaborative.
- Work with commercial colleagues from the start.
- Have the skills and expertise needed to implement and use AI.
- Use these principles alongside your organisation’s policies and have the right assurance in place.
The playbook points out the limitations of AI in the form of possible bias, its reliance on the quantity and quality of data, the difficulties in producing a system that is 100% accurate under all conditions, and the cost and sustainability.
It adds that generative AI – with which there is now a large amount of experimentation in the public sector – has limitations such as a risk of creating content that is factually incorrect, a lack of critical thinking, sensitive or inappropriate context, an absence of domain expertise and the fact that many models do not have real time access to data.
There are also sections on the legal considerations, data protection and privacy, security, governance, procurement and working with companies on the deployment of the technology.
Part of surge
Publication of the document is the latest step in the surge of activity to boost the role of AI in public services. It has come soon after the Government published an AI Opportunities Action Plan to promote the use of the technology in public services, and the publication of its Blueprint for modern digital government, in which the development of new solutions based on AI plays a central role.
DSIT also highlighted a number of examples from the Algorithmic Transparency Records of how AI and algorithmic tools are being used to speed up decision making and improve public services.
They include weather and climate forecasting by the Met Office, habitat mapping by Natural England, MOT risk rating by the Driver and Vehicle Standards Agency, and the Effective Proposal Framework for probation practitioners by the Ministry of Justice.
