Using AI as a Nonprofit Consultant: What to Say When Clients Ask

Your client just asked whether you use AI.

You use a note-taking tool during calls. You used AI to help structure that last report. You're not doing anything wrong - but you're standing in front of a question you haven't fully prepared for, because nobody gave you a script for this.

This is where nonprofit consultants are right now. The sector is catching up to AI in an uneven, sometimes contradictory way. Organisations have policies that say no while their staff quietly uses AI tools anyway. Leadership teams are saying "we need to figure this out" while smaller teams are already adopting tools with AI built in. And consultants are in a client landscape where the question isn't whether AI is involved in your work - it's whether you're ready to talk about it.

Valerie Ehrlich, PhD, founder of Mission Bloom, has spent two decades working in nonprofit learning, evaluation, and organisational strategy. Her work now focuses on responsible AI adoption in the social sector. In this Fracture episode, she and Cindy Wagman work through the real questions nonprofit consultants are facing.

The short answer: Clients are starting to ask about AI, and consultants need to be ready. The key distinctions are proactive vs reactive disclosure, and client-facing vs backend AI use. Being transparent about your process - while grounding your value in the expertise you bring - is how you keep client trust without lowering your fees.

🎧 Listen to the full episode on Apple

🎧 Listen to the full episode on Spotify

▶️ Watch on YouTube

What Is the Temperature on AI in the Nonprofit Sector Right Now?

It's mixed.

"There's no single consensus," Valerie says. "There are hot and cold pockets across organisations and even within them." Some organisations have official policies banning AI use by contractors. At the same time, AI is already built into tools those same organisations use daily - Google Search, document editors, email platforms. "Organisations don't even realise they're using AI because it's now embedded in tools like Google Search."

There are also what Valerie calls shadow systems: situations where official policy says one thing and people are quietly doing something else. This matters for consultants because your client's stated position on AI may not reflect what's actually happening inside their organisation.

The takeaway: don't assume the stated policy is the whole picture. Be prepared for both ends of the spectrum.

Should Consultants Be Proactive or Reactive About AI Disclosure?

Both approaches are valid. What's not valid is being caught off guard.

Valerie frames it as a choice about positioning: some consultants define their approach upfront and advocate for it; others educate clients about AI as part of the engagement. If you're proactive, you explain your AI use in your onboarding conversation and build it into your contract. If you're reactive, you have a clear, confident answer ready when clients ask.

The questions clients are asking are getting specific: What happens to our data? What tools are you using, and what are their security settings? Would you be willing to use our tools instead? These questions have real implications for your pricing and scope.

Being transparent also addresses the unspoken concern behind most of these questions: whether your deliverables are AI-generated and what your expertise actually contributes. Explaining where AI supports your process and clarifying that deliverables are always reviewed and refined by you addresses that concern directly.

What Is the Difference Between Client-Facing and Backend AI Use?

This distinction does a lot of work, and it's underused.

Client-facing AI use - using AI to generate or shape deliverables - is where transparency is most critical. If a report, strategy document, or training design was substantially AI-generated, your client may reasonably want to know.

Backend AI use - using AI to manage your own workflow, summarise your notes, track project history, or think through a complex problem - is different. Valerie's example is using AI as a "second brain": tracking conversations and context to stay organised and fully present with clients. "Using a note-taking tool during meetings can help you stay fully present rather than distracted by writing notes." This kind of backend use doesn't directly affect client deliverables, and the disclosure expectations around it are different.

Start by identifying your bottlenecks - where you lose time, where you get stuck - and explore how AI might support those areas specifically.

"The real value consultants bring is relational - how we connect, interpret, and apply expertise. AI can support that by freeing up time or improving preparation." - Valerie Ehrlich, PhD

What Consultants Need to Know About AI and Data Privacy

This is where consultants get tripped up - not because they're careless, but because the defaults on most tools are not set up with confidentiality in mind.

A few points Valerie is direct about:

Never record a client call without consent. Even where it's technically legal, it's a transparency issue that can erode trust quickly.

Don't enter sensitive or personally identifiable client information into AI tools unless you fully understand how that data is handled. This matters especially for free-tier tools.

Pay for tools rather than relying on free versions. Free tools often monetise your data. If you're using AI in your client work, treat it as a business cost.

Read the terms of service and check your tool settings regularly. "Policies can change frequently. Even subtle changes can have major implications for how your data is used."

This is not a reason to avoid AI. It's a reason to treat AI tools the same way you'd treat any tool that touches client data - with careful evaluation and regular review.

🎧 Listen to the full episode on Apple

🎧 Listen to the full episode on Spotify

▶️ Watch on YouTube

Frequently Asked Questions

Do I have to tell clients I use AI?

The expectation around disclosure is evolving. If AI is shaping your client deliverables, transparency is warranted. If you're using AI for your own backend workflow - organising your notes, summarising your own thinking - the threshold for disclosure is lower. Either way, being ready to answer the question confidently is more important than having a universal policy.

What should I say when a client asks if I use AI?

Know what tools you use, how data is handled, and what role AI plays in your process. Explain it clearly: what AI does in your workflow, what you review and refine personally, and how your expertise shapes the output. You're not defending yourself - you're explaining your process. That's a professional conversation, not an uncomfortable one.

What are shadow systems and why do they matter for consultants?

Shadow systems are situations where an organisation has an official AI policy (often restrictive) but staff are using AI tools anyway - sometimes without realising it, because AI is built into standard tools. Your client's stated AI policy may not reflect what's actually happening inside, which means the conversation about AI use is more complex than a simple yes/no.

Will AI lower my consulting fees?

Some clients will raise this. Valerie's response: the quality of AI output depends on the expertise of the person using it. A nonprofit consultant with deep sector knowledge will get far better results from AI tools than someone without that knowledge. AI doesn't replicate expertise - it works better when expertise guides it. That's your value proposition, not a threat to it.

Is it okay to use AI note-taking tools during client meetings?

With consent, yes. Never record any conversation without telling the other party. Many consultants find that transparent use of note-taking tools builds trust, because it lets them stay fully present in the conversation rather than managing notes simultaneously.

What AI tools should nonprofit consultants be using?

Rather than recommending specific tools - which change quickly - the better question is: where are your bottlenecks? What takes too long? What do you get stuck on? Look for tools that address those specific areas. Start with paid versions with clear privacy policies and read the terms of service before putting any client data in.

How do I stay current on AI without it taking over my practice?

Valerie describes the goal as "engaged skepticism" - not blind enthusiasm, not wholesale resistance. Use what helps you. Question what doesn't. Review your tools regularly as policies and capabilities change. Your core values and practices as a consultant remain constant; AI is a tool that works within them.

The conversation about AI in your consulting practice is not going away. Starting from your values, being transparent with clients, and staying grounded in the relational expertise you bring - that's the approach that serves you and your clients well over time.

🎧 Listen to the full episode on Apple

🎧 Listen to the full episode on Spotify

▶️ Watch on YouTube