Here's the thing about trust: you can spend years building it and lose it in seconds. And right now, the nonprofit sector is standing at a crossroads that could determine whether AI becomes our greatest ally or our most dangerous liability.

I've been thinking about this a lot lately, especially after diving into Bonterra's recent report that reveals a stark contradiction in our sector. Ninety-two percent of nonprofit and CSR leaders see AI's potential to transform our work. **But here's the kicker: nearly the same percentage are deeply concerned about data protection and ethics. **

This isn't just numbers on a page. It's a crisis of trust waiting to happen.

The stakes couldn't be higher

Here’s what this looks like in my work at Zero Abuse Project: I build AI-driven tools that help our team manage complex programs and train child-protection professionals who handle some of the toughest cases out there. The data involved is deeply sensitive—victim disclosures, investigative notes, trauma histories. So when I’m designing these systems, I’m not just thinking about workflow or UX. I’m thinking about the child behind every line of information. Their safety. Their family. Their future.

This is the reality for nonprofits everywhere. We don't just handle data; we hold stories. We carry the weight of human vulnerability in every spreadsheet, every client record, every intake form. Health information, financial hardship, immigration status, domestic violence histories. This isn't abstract information. These are real people who trust us with their most private struggles.

And yet, we're being asked to hand this sacred trust over to AI systems that often operate like black boxes.

The trust deficit is real

Here's what keeps me up at night: I see nonprofits rushing toward AI adoption because the pressure is immense. Funders want to see innovation. Boards want efficiency. Staff are drowning in administrative tasks. The promise of AI solving our resource constraints is intoxicating.

But in our eagerness to embrace these solutions, are we asking the right questions? When a vendor promises their AI will "streamline your client management," do we know where that data goes? How it's stored? What happens if the company gets acquired? If they change their privacy policy?

I've sat in countless nonprofit webinars where leaders nod enthusiastically about AI implementation, then ask afterward, "But how do we actually know it's safe?" That gap between excitement and understanding is where trust erodes.

The Bonterra report confirms what I'm seeing in the field. We want the benefits, but we're terrified of the risks. And honestly? We should be.

What ethical AI actually looks like in practice

At Lilac Creative, when we help nonprofits evaluate AI tools, we don't start with features and benefits. We start with values alignment. Does this tool respect the dignity of the people whose data it processes? Can we explain in plain language how it works? Would we be comfortable if our beneficiaries knew exactly how their information was being used?

These aren't just nice-to-have questions. They're ethical imperatives.

I've learned this through trial and error. When we launched the NOVA Tool, we didn't just focus on the technology. We built transparency into every layer. Officers know exactly what data is collected, how it's used, and how it's protected. They can see the reasoning behind every recommendation. Human oversight isn't optional; it's built into the workflow.

This approach takes longer. It's more expensive. But here's what we discovered: when people understand and trust the system, adoption skyrockets. When they don't, even the most brilliant AI sits unused.

The path forward isn't about choosing sides

The solution isn't to halt AI adoption. That ship has sailed, and frankly, the potential for good is too significant to ignore. When I see how AI-powered simulations can help officers approach survivor interviews with more empathy and precision, when I watch case management tools free up social workers to spend actual time with clients instead of paperwork, I know we're onto something transformative.

But transformation without trust is just disruption. And disruption in the nonprofit space means real people get hurt.

We need to demand better from our vendors.

Plain-language explanations shouldn't be a luxury; they should be standard.

When I'm evaluating an AI tool for Zero Abuse Project, I don't just ask for technical specifications. I ask them to explain their data practices like they're talking to my neighbor. If they can't do that, we keep looking.

We need to mandate human review for decisions that impact beneficiaries. AI can inform our choices, but it can't make them for us. The moment we remove human judgment from decisions about human lives, we've crossed a line.

And we need to prioritize secure, governed platforms over convenient public tools. Yes, ChatGPT might seem like an easy solution for drafting grant proposals, but if you're inputting client information or strategic planning details, you've just made that data part of a training set.

Is that a risk you're willing to take?

Building on bedrock

As I prepare for 2026, I'm not going to train myself on the latest and greatest algorithms. I'm going to find more ways to advocate for a different conversation entirely. One where ethical data practices aren't an afterthought but the foundation everything else is built on.

Because here's what I know from years of working at this intersection: ***AI's potential for social good is directly proportional to our commitment to ethical stewardship. ***The most powerful tool in the world is worthless if the people it's meant to serve don't trust it.

The nonprofit sector has always been built on relationships, on trust, on the belief that we can create change by putting people first. AI doesn't have to change that. In fact, when implemented ethically, it can amplify those very values.

But only if we refuse to sacrifice trust for efficiency. Only if we demand transparency even when it's inconvenient. Only if we remember that behind every data point is a human being who deserves our very best judgment.

The choice is ours. And the time to make it is now.