Use of AI
How charities can create real value with AI
A framework for working out where AI actually helps — and where the conversation in the sector keeps getting stuck.
As AI develops, I've been making a proactive effort to shift from "play and learn" to "add real value". Many conversations in the sector focus on automation or trust concerns, but this overlooks the true potential of AI to help people.
I've been developing a framework that I think gives a cleaner picture of where AI actually creates value. It has two dimensions. The first is the type of value. The second is who captures it.
Three types of value
AI creates value in three fundamentally different ways.
Scale
Doing more of the same, faster and cheaper. More content produced, more people reached, less time spent on tasks that eat up whole afternoons. This is where most organisations start, and that makes sense — the wins are quick and visible.
Lift
Doing things better than before. Not just faster — actually higher quality. Better decisions because you’ve analysed more data. Sharper strategy because you’ve had a thinking partner push back on your assumptions. Deeper research because you can synthesise across sources that would take weeks to read properly.
Unlock
Doing things that simply weren’t possible before. Not faster. Not better. Genuinely new. Personalisation at scale. Democratised expertise. Capability that previously required a team of twenty, now available to an organisation of five.
Scale and Lift are improvements on existing activity. Unlock is a different category entirely.
Who captures the value?
The second dimension is about who benefits, and this is where it gets interesting for charities specifically.
Value flows to individuals, organisations, and society. Put the two dimensions together and you get this:
| Scale | Lift | Unlock | |
|---|---|---|---|
| Individual | Saves time | Better decisions | New capabilities |
| Organisation | Saves money | Higher impact | New models |
| Society | Reaches more people | Better outcomes | Democratised access |
Most charity AI conversation lives in the top-left of this table. The really important territory — for us — is the bottom-right. Some examples:
| Scale | Lift | Unlock | |
|---|---|---|---|
| Individual | Fundraiser uses AI to draft 50 donor thank-you letters in the time it used to take to write 5 | Comms manager uses AI to stress-test a campaign message before it goes out, catching tone issues a tired team missed | A small charity CEO gets CFO-level financial modelling they could never have afforded from a consultant |
| Organisation | A charity replaces an outsourced data processing contract worth £30k a year by automating the work internally with AI | A hospice uses AI to analyse supporter feedback across 3 years and spots patterns that reshape their stewardship strategy | A regional food bank builds a personalised supporter journey for 10,000 donors — something that previously required a team of 20 |
| Society | A mental health charity reaches 10x more people with its awareness content at no extra cost | A disability charity uses AI research synthesis to build a stronger evidence base for policy advocacy | A legal aid charity gives free, accurate legal guidance to people who previously had no access to any professional support |
Naturally, there will be overlaps. The idea behind this article originally stemmed from someone in my team who used Copilot to teach themselves how to create a PowerBI dashboard that integrated different data sources, allowing us to see which leads converted better and segment our comms — Unlocking new skills for them, Lifting the quality of our campaigns, and Scaling our use of insight.
Why do this?
I find it helpful, as I try and understand this fast changing world and technology, to come back to why pursue this. One in ten people live on less than $3 a day. 300,000 people drown every year. A million species face extinction. A civilian dies in armed conflict every 12 minutes. These, and every other cause we care deeply about, are not problems without solutions and we must use all the tools available to us. AI can be a negative force or a positive one, I want our sector to use it for good.
Where does this leave us?
Most charities will start with Scale. That's fine — the quick wins build confidence and free up capacity.
But the organisations that will pull ahead over the next five years won't just be the ones that automate fastest. They'll be the ones that ask what becomes possible that wasn't before — and build deliberately towards it.
The question I'd encourage every charity to sit with isn't "how do we use AI to do what we already do more cheaply?" It's "what could we do for our beneficiaries that we've never been able to do before?"
That's the Unlock question. And I think it's the most important one in our sector right now.
What are the Unlock moments you're seeing in your work?
Steve