Using AI responsibly in your faith-based organisation: what it means for the people you serve

Artificial intelligence has quietly arrived in the day-to-day life of faith-based organisations. It drafts newsletters, tidies up grant applications, answers admin emails, summarises documents. The question facing trustees is no longer whether AI is used in their organisation, but how — and whether that use is consistent with their duty to the people they serve.

This article isn't an abstract discussion of "AI and faith". It's about something more practical and more important: when your organisation uses these tools, what happens to the vulnerable people whose information and trust you hold? That's the lens that matters.

Start from your duty, not the technology

Faith-based organisations exist to serve — often to serve people at their most vulnerable. Everything about how you adopt AI should flow from that. The useful question is not "is AI good or bad?" but "does this use of AI honour our duty to the people who depend on us?" Held that way, a few clear principles follow.

Principle 1: A human stays responsible

AI can assist; it cannot be accountable. A tool can draft a letter, suggest a response or summarise a case — but a person must review it, exercise judgement, and own the outcome. This matters especially where decisions affect people: who receives help, how a safeguarding concern is handled, what's said to someone in distress. Never let a tool make, or appear to make, a decision about a person without a human meaningfully in the loop.

For trustees, this is simply your existing duty of care in a new setting: you remain responsible for what your organisation does, whoever — or whatever — helped produce it.

Principle 2: Protect people's data — especially the most sensitive

This is where faith organisations need to be particularly alert. Under UK data protection law (the UK GDPR and Data Protection Act 2018), some information is "special category data" — treated as more sensitive and given extra protection. It includes data about health, sexual orientation, ethnicity — and religious belief itself.

That has a striking implication: your records about who your members and beneficiaries are may already be special-category data, simply because they reveal religious affiliation. Add the health, family or hardship details a faith charity often holds in the course of pastoral care or welfare support, and you are routinely handling some of the most protected categories of personal information there are.

So, before you put any real person's information into an AI tool:

• Don't paste sensitive personal details about beneficiaries, members or service users into public AI tools you don't control. Once it's in, you may have no idea where it goes.
• Understand where the data goes. Free consumer tools may use what you enter. Know whether the tool retains, trains on, or shares your inputs.
• Minimise. If a tool genuinely helps with a task, can you use it with the personal details stripped out or anonymised?
• Know your obligations. Significant new uses of personal data may require a Data Protection Impact Assessment; the people whose data you hold have rights over it. The ICO publishes practical guidance for organisations of every size.

Principle 3: Safeguarding doesn't pause for new technology

Everything your safeguarding policy stands for applies to AI use too. Feeding a vulnerable person's circumstances into an uncontrolled tool is a safeguarding risk, not just a data one. A photograph, a disclosure, a child's details, an account of someone's hardship — these demand the same protectiveness in an AI context as anywhere else. Build AI into your safeguarding thinking rather than treating it as a separate, technical question. (See our guide to keeping your safeguarding robust.)

Principle 4: Be honest about how you use it

Authenticity matters in faith communities, and it matters to funders. Two areas deserve particular care:

• Grant applications and reporting. Using AI to help draft or polish an application is not wrong in itself — but the claims must be true and the voice must be genuinely yours. Never let a tool generate impressive-sounding statements that misrepresent your work. Funders increasingly ask about AI use; honesty protects your reputation and relationships.
• Pastoral and community communications. A message of comfort, condolence or spiritual encouragement carries weight precisely because a human means it. Be thoughtful about where AI assistance is appropriate and where it would hollow out something that should be personal.

Transparency about your own AI use — being willing to say where and how you use it — is itself a mark of integrity.

Principle 5: Watch for bias and exclusion

AI systems can reflect and amplify bias, and they can quietly exclude. Be alert to whether a tool serves everyone you serve — across language, age, disability and digital confidence — or whether leaning on it widens the gap for those already on the margins. The people most likely to be disadvantaged by a poorly chosen tool are often exactly the people a faith charity most wants to reach.

A simple "before you use a tool" check for trustees

Before adopting an AI tool in your organisation, ask:

1. Whose data might go into it, and how sensitive is it?
2. Where does that data go, and is that acceptable?
3. Who reviews the output before it affects a person or goes out in our name?
4. Is it honest — would we be comfortable explaining this use to our members and funders?
5. Does it include or exclude the people we serve?

If you can answer those five clearly and comfortably, you're likely using AI in a way that's consistent with your values. If you can't, that's the signal to pause.

The bottom line

AI can genuinely lighten the administrative load on stretched faith organisations and free up time for the work that matters. Used carelessly, though, it can put the dignity, privacy and safety of vulnerable people at risk. The difference lies not in the technology but in the discipline around it — keeping a human responsible, guarding people's data, honouring safeguarding, staying honest, and including everyone. That's not a brake on using AI well. It's what using it well actually means.

---

This article is general information, not advice. Rules and technology change quickly and can depend on your circumstances. Check the current position with the Information Commissioner's Office (ICO), your safeguarding lead, or get in touch and we'll help.

Sources verified (June 2026):

• ICO, Guidance on AI and data protection — https://ico.org.uk/for-organisations/uk-gdpr-guidance-and-resources/artificial-intelligence/guidance-on-ai-and-data-protection/
• ICO, What is special category data? (religious belief as special category data) — https://ico.org.uk/for-organisations/uk-gdpr-guidance-and-resources/lawful-basis/special-category-data/what-is-special-category-data/
• ICO, AI and data protection risk toolkit — https://ico.org.uk/for-organisations/uk-gdpr-guidance-and-resources/artificial-intelligence/guidance-on-ai-and-data-protection/ai-and-data-protection-risk-toolkit/
• The Data (Use and Access) Act 2026 (Royal Assent 19 June 2026) amends the UK GDPR/DPA 2018 — check the ICO site for current guidance.