AI has moved very quickly.

What started as experimentation with tools like ChatGPT has become something much bigger. AI is increasingly being built into the software charities already use, while staff and volunteers are experimenting with tools themselves.

That creates an interesting challenge.

AI adoption doesn't necessarily start when leadership decides to adopt AI.

It may already be happening.

Someone is using it to draft a funding application. Someone is summarising a meeting. Someone is analysing a spreadsheet. Someone is creating communications. Someone has discovered an AI feature inside an existing platform.

The question for charity leaders is therefore no longer simply:

“Should we be using AI?”

It's:

“What does AI actually mean for our organisation, and how do we make sure we're using it to create genuine value?”

After several years working hands-on with charities across digital transformation and AI adoption, I've become increasingly convinced that the organisations that get this right won't necessarily be the ones moving fastest.

They'll be the ones asking better questions.

Stop starting with the technology

There is a lot of noise around AI.

New models. New agents. New automation tools. New promises of productivity. It's understandable that organisations feel pressure to keep up.

But I think this creates a dangerous temptation to start with the technology.

“Should we use ChatGPT?”

“Should we buy Copilot?”

“Could we build an AI agent?”

These are useful questions eventually. They're just not the first questions I'd ask. I'd start somewhere much simpler:

Where are we experiencing friction, inefficiency or missed opportunity?

• Perhaps your fundraising team spends hundreds of hours on repetitive research.
• Perhaps valuable information is sitting unused in your CRM.
• Perhaps your teams are manually producing reports from several different systems.
• Perhaps your supporter journeys are too generic because people don't have the time or insight to personalise them.
• Perhaps highly skilled people are spending too much of their working day searching, summarising, copying and moving information around.

Those aren't AI problems. They're organisational problems that AI may be able to help solve. That distinction matters.

The three opportunities I think charities should be exploring

1. Create capacity

Efficiency is an obvious opportunity for AI. But I'd challenge how we define productivity. The objective shouldn't be:

“How can we make our people do more?”

It should be:

“What could our people do with the time we give back?”

That might mean:

• More time with supporters
• More time developing relationships
• More time with beneficiaries
• More time thinking strategically
• More time doing work that requires judgement and empathy

The best efficiency gains aren't necessarily about making people work faster.

They're about removing work that doesn't need to consume human capacity in the first place.

That's a very different proposition.

2. Strengthen income generation

This is an area where I think the charity sector has significant untapped potential.

Many organisations already hold enormous amounts of information about their supporters, donors, prospects and fundraising activity.

The question is whether they're turning that information into useful insight.

AI could potentially support:

• Better segmentation
• Propensity modelling
• Prospect identification
• Fundraising research
• Personalisation
• Campaign optimisation
• Supporter retention
• More relevant supporter journeys

The technology itself isn't the interesting part.

The interesting part is what happens when better data and better insight lead to better decisions.

If AI helps an organisation understand who is most likely to respond, what they might respond to, when they might respond and how best to engage them, the potential value goes well beyond saving a few hours.

It could mean stronger, more sustainable income for the mission.

3. Improve the supporter experience

There is another opportunity that I think deserves far more attention.

AI should not only make the organisation more efficient.

It should make the experience of engaging with the organisation better.

Think about:

• More relevant communications
• Faster responses
• Better understanding of supporter needs
• Less repetition
• More joined-up journeys
• Earlier identification of disengagement

Used well, AI doesn't have to make an organisation feel less human.

Quite the opposite.

The best use of AI should give people more capacity to be human.

If we're simply using AI to produce more generic communications at greater speed, we've potentially automated the wrong thing.

If we're using it to understand people better and remove friction from their experience, that's much more interesting.

But here's the uncomfortable bit

The opportunity is significant. So is the responsibility. Purpose-led organisations hold some of the most sensitive information imaginable.

Beneficiary information.

Donor information.

Health information.

Safeguarding information.

Personal stories that people have trusted an organisation to handle carefully. That means AI adoption cannot simply be treated as another productivity initiative.

Trust is part of the technology architecture.

If someone enters sensitive information into an AI tool, leaders need to understand:

• Where does that information go?
• Who can access it?
• What are the provider's terms?
• Is the information retained?
• What happens to it afterwards?
• How are AI outputs checked?
• Who is accountable when something goes wrong?

And this isn't necessarily about someone making a reckless decision.

The bigger risk can be AI adoption happening quietly and informally without anyone really knowing what's going on.

As I explored in my recent Safe & Ethical AI Adoption Guide for Purpose-Led Organisations, the starting point isn't having every answer.

It's having an honest picture of where you are today.

Your AI strategy might be the wrong place to start

I don't believe every charity needs to rush into a huge AI transformation programme. Nor do I think every organisation needs a 50-page AI strategy. Before strategy, I'd want to understand the organisation.

What is already happening?

Where are the opportunities?

What are the risks?

What does the data look like?

How confident are the people?

• What systems are already in place?

Where could AI create meaningful value?

And where shouldn't AI be used?

That's the groundwork.

Only then can you make sensible decisions about investment, technology and priorities.

Technology should support the strategy (It shouldn't become the strategy).

AI readiness isn't about being "good at AI"

An organisation isn't AI-ready because everyone has a ChatGPT account. It isn't AI-ready because you've appointed an AI champion. It isn't AI-ready because the CEO attended an AI conference. And it isn't AI-ready because somebody has written an AI policy.

Readiness is much more nuanced.

It's about having enough of the right foundations in place to make good decisions about AI.

That means looking at:

• Governance
• Data
• People
• Culture
• Technology
• Opportunity
• Risk
• Organisational appetite for change

The guide I've created deliberately starts from the principle that you don't need every answer today.

What matters is having an honest, shared picture of where you stand now.

And that last word matters.

Shared.

One person's view isn't enough

Ask the CEO whether the organisation is ready for AI and you may get one answer.

Ask IT and you may get another.

Ask fundraising and you could get something completely different.

Ask frontline teams and you might discover that AI is already being used in ways leadership didn't know about.

That's why I don't think AI readiness should sit with one person or one department.

It needs an organisational view.

Not because everyone needs to agree ,quite the opposite.The differences in perception can be some of the most valuable information you uncover.

If leadership thinks you're ready but staff don't feel confident, that's important. If staff are experimenting rapidly but governance hasn't caught up, that's important.

If everyone wants to use AI but the underlying data isn't reliable, that's important. The gaps are where the work begins.

So what should charity leaders do now?

Not everything, But something. I'd start with five questions:

1. What is already happening?

Do we know where AI is being used across our organisation?

2. Where could it create genuine value?

Where could we create capacity, strengthen income or improve experiences?

3. What could go wrong?

Where are our greatest risks around data, privacy, bias, security and accountability?

4. Are we ready?

Do our people, data, governance and technology provide a sensible foundation?

5. What should we do first?

Not everything. The things most likely to create meaningful value safely. AI adoption doesn't have to start with a huge transformation programme. It can start with:

• One problem
• One process
• One team
• One well-chosen use case

But it should start deliberately.

The charities that benefit most from AI won't necessarily be the ones using the most AI

I think we're going to see a lot of noise over the next few years. Some organisations will announce dozens of AI initiatives. Some will build impressive demonstrations. Some will spend significant amounts of money. Others will quietly use AI to:

• Remove friction
• Improve decisions
• Strengthen income
• Improve supporter experiences
• Give their people more time to focus on their mission

I'd put my money on the second group. Because ultimately AI isn't the outcome.

Better services are the outcome.

Greater impact is the outcome.

Stronger relationships are the outcome.

More sustainable income is the outcome.

More capacity for the people doing important work is the outcome.

AI is simply one of the most powerful tools we've had to help us get there. But only if we use it with intent.

Purpose first. People always. Technology with intent.

That's the principle behind Digi4Good.

And it's why I've created the free Safe & Ethical AI Adoption Guide for Purpose-Led Organisations.

If you're trying to work out what AI should actually mean for your charity, start by cutting through the noise and understanding where you stand.

Download the free guide

Not sure is a completely normal place to start.