This is a curated library of practical resources, official guidance, and key context. It is designed to help you find one useful thing, use it, and move on with your work.
These resources are a practical starting point. They are not legal, data protection or safeguarding advice. Seek specialist advice for your organisation’s specific circumstances.
Click the boxes to expand them.
A straightforward introduction to generative AI for charity staff and volunteers. It explains enough to help someone use a tool sensibly, ask better questions and know when to stop and check, without expecting them to become an expert.
A quick decision aid for the moment before someone pastes information into an AI tool. It covers personal data, special-category and sensitive information, beneficiary and safeguarding information, confidential organisational material, commercial information, unpublished funding information, information about other people, and practical anonymisation.
A set of adaptable prompts for real non-profit work, including planning, drafting, meetings, communications, research and leadership tasks. It should teach people how to adjust a prompt for their own context, rather than treating prompts as magic words.
Ten practical questions for trustees. This is about good judgement, accountability and the questions a board needs to ask. Trustees do not need technical expertise to do their job well.
A realistic list of lower-risk, useful tasks that can save time around the work. Examples should include turning rough notes into a first draft, improving structure and clarity, preparing an agenda, summarising public information, planning a workshop, and turning one piece of content into several formats. It must say clearly that AI is not right for every task.
A practical guide to using AI thoughtfully when it adds genuine value. It covers using AI on purpose, giving it a clear brief, using text where text will do, asking for less, and asking providers better questions about energy, water and sustainability.
A ready-to-adapt message that gives staff permission to start small, sets simple boundaries for using AI and makes clear where to ask questions. It should be non-punitive and non-alarmist.
A simple checklist to help a charity decide whether an AI tool is worth testing. It covers the problem, beneficiaries, whether AI is the right solution, expected frequency of use, data, the impact of errors, supplier data practices, connected systems, cost and training, human checking, export and exit, supplier or product changes, and how the charity will know whether the tool helped.
A guide for anyone considering an AI agent or automated workflow. It covers access, least privilege, whether the agent can read, change, draft, send or publish, logging, external communications, human approval for consequential actions, stopping, escalation, acceptable failure and independent checking.
Click the boxes to expand them.
The Charity Commission is clear that AI does not remove or dilute trustee duties. Trustees remain responsible for decisions, including where an AI tool has supported the work. Use this as the starting point for governance conversations, alongside the charity’s own purposes, duties and risk management.
The Commission’s AI material is a blog post rather than a dedicated statutory AI guide. It was checked on 19 September 2026 and should be reviewed again before publication.
The Fundraising Regulator’s guidance is an important reference point for any charity using or advising on AI in fundraising. It asks organisations to remain responsible and accountable for AI outputs. It covers inaccurate or generated information, hallucinations, bias, personal and non-public data, data storage, transparency and human oversight.
The page was last updated on 8 December 2025 and was checked on 19 September 2026.
The ICO is the UK data protection regulator. Its AI resources matter wherever personal data, sensitive information, profiling, automated decision-making or high-risk processing is involved.
The ICO is updating its automated decision-making and profiling guidance following the Data (Use and Access) Act 2025. The consultation closed in May 2026 and the ICO lists final guidance for winter 2026. Label this accurately as an update in progress, not current final guidance. View ICO tech horizons.
The UK does not have one single, overarching AI law. Existing legal duties and regulators apply according to the context, particularly where AI affects recruitment, assessment, eligibility, prioritisation, risk or access to services.
The toolkit is a useful starting point for people designing, operating, procuring or delivering AI-enabled products. It was published on 8 September 2026.
This Joint Committee on Human Rights report is useful context for consequential decisions about people. It makes recommendations to government. It is not current law or government policy.
These are useful perspectives from organisations working with charity leaders and smaller charities. They can help you think through where AI might be useful in everyday work, alongside the official guidance on governance, fundraising and data protection.
A practical introduction for smaller charities that are working out where to begin. It covers everyday uses of AI across communications, fundraising, governance, planning and evaluation, while recognising the risks around data protection, cyber security and misinformation. The page was reviewed by NCVO in August 2026.
A leadership-focused introduction to how AI could support administration, fundraising and impact measurement. It also raises useful questions about privacy, security, bias, human judgement and the information that should never be put into an AI tool. This is a guest article by Qlic IT for Charities, published by ACEVO.
AI agents can work through a series of steps towards a goal, including searching files, using connected tools or taking actions. The question is not only whether an agent can do the work. It is what it can access, change or send, and what happens when it gets something wrong.
Start with the AI for Non-Profits guide, Before You Give AI Permission to Act. Use external material when a team needs more detail.
This is a strong practical source on limiting access, starting small and planning for failure. It is particularly useful where an agent would have access to sensitive or critical systems.
The ICO is developing dedicated guidance on agentic AI. Its planning page says a public consultation is due in September 2026, with final guidance expected in spring 2027. Treat this as developing guidance.
AI can reflect and amplify existing inequalities, especially where tools classify people, prioritise support, assess risk, moderate content, screen applications or make recommendations about people. This is not a separate equality add-on. It belongs in decisions about whether and how to use AI.
A current framework on representation, safety, data protection, moderation and meaningful involvement of LGBTQ people.
A foundational report on how technical systems can reproduce inequalities connected to gender, race and power.
A UK example of the risks of predictive systems, including biased data, lack of transparency and automation bias.
A current UK case study on transparency, accuracy, procedural fairness and the impact of AI-supported decisions on people seeking asylum.
Use this as a starting point for thinking about where data gaps make supposedly neutral systems less reliable for particular groups.
AI raises particular questions for organisations working with children and young people, including safeguarding, privacy, identity, relationships, AI companions, manipulated images and age-appropriate AI literacy.
Led by Nicki Watts, AI Youth explores generative AI and young people’s wellbeing, identity, relationships, agency, safety, emotional wellbeing and critical literacy. Its participatory approach is particularly useful for charities wanting to work with young people rather than simply designing around them.
A focused resource on manipulated images and how organisations can respond. It is more useful here than a generic list of school resources.
AI can be useful. It also has an environmental cost. There is no settled calculator for the environmental impact of everyday charity AI use. Impact depends on model, task, data-centre operations, electricity, water, hardware, supplier and the total scale of use.
Start with the AI for Non-Profits resource, Using AI thoughtfully: the environmental impact. It gives people sensible practical actions without pretending every prompt has a fixed footprint.
A useful source on why measuring AI’s environmental impact is difficult and why consistent methods matter.
A broad, evidence-led report on data centres, electricity demand, energy supply and the potential role of AI in energy systems.
Useful UK context on the infrastructure and public policy behind growing computing capacity. It is not a charity sustainability guide.
Research on where AI could support emissions reduction, alongside the need to account for AI’s own footprint.
