The AI wave in advancement technology arrived with impressive velocity. Case studies are circulating about AI-generated wealth screens, donor sentiment analysis, and automated stewardship workflows. At conferences, slide decks showcase prototypes of predictive prospecting engines, and headlines suggest AI is the long-awaited unlock for fundraising productivity. To be clear, they are right.

But in the background of this movement, practitioners are asking many legitimate questions: Is it ready? Will it work on our data? Can we support it with our current teams and systems? Is it solving our actual problems, or is it just new and shiny?

These aren’t the questions of Luddites or the change averse. They are the questions of what we will call the AI pragmatist.

What is an AI Pragmatist?

An AI pragmatist doesn’t resist innovation. Rather, they resist the illusion that AI saturation is simply inevitable and inherently good. They are not cynics, but neither are they cheerleaders. Pragmatists evaluate tools not by their trendiness or technical novelty, but by their usefulness to organizations that operate outside a sandbox with real constraints: budget, staff, risk tolerance, data messiness, and cultural adoption.

They understand something that’s often understated in these presentations, particularly in fundraising: mistakes aren’t cheap. Donor relationships != API calls. Misclassifying a major gift prospect, misinterpreting intent from natural language processing, or sending a stewardship message to the wrong constituent erodes trust. And trust is advancement’s true currency.

AI evangelists can innocently frame this tension as a lack of imagination or risk appetite. But the AI pragmatist knows the opposite is true: strategic restraint and saying “not yet” is often a form of leadership and choosing thoughtful timing and practical execution is not a lack of ambition.

Slow is Fast, Fast is Smooth

Many AI vendors in the advancement space are moving fast because they have to. The market rewards flash. It’s easier to demo an AI-powered engagement dashboard than to replatform decades of technical debt. It’s easier to deliver sentiment analysis than to codify your data definitions.

But advancement leaders live with the consequences of every shortcut. If your team can’t govern your current CRM data model, layering a rogue black-box on top of it won’t solve the problem. Every point of automation without a solid foundation becomes a risk multiplier.

AI pragmatists ask a different set of questions: “What must be true for this to work? Operationally, ethically, and organizationally?” They take the time to make those conditions real: improving data hygiene, establishing clear governance, understanding business logic, documenting naming conventions, and modernizing reporting pipelines so humans can still see what the machine thinks.

Restraint as a Form of Leadership

Being a pragmatist doesn’t mean sitting AI out until it reaches full maturity. On the contrary, it means investing now in the areas that will make it successful later:

  • Structuring and cleaning your data to make future training sets reliable and traceable. Ever heard of garbage in, garbage out?
  • Logging user behavior in a way that will one day inform intelligent workflows
  • Building organizational muscles for ethical review, feedback loops, and data strategy
  • Clarifying what decisions and outcomes matter most, and executing where AI makes sense for your organization

This is what it means to be AI-ready in advancement: not to paste a donor’s info into ChatGPT to get a thank you letter, but to design the systems, context, and judgment that AI will serve.

Flash Can’t Fix the Fundamentals

The challenge for advancement tech vendors and leaders is that a pragmatic posture doesn’t always yield immediate accolades. It doesn’t produce a flashy demo. It doesn’t get quoted in keynotes. But it builds organizational capacity, strategy and governance. And those are what sustains innovation long after the hype fades.

I’ve helped institutions implement CRMs, integrate gift systems, modernize data pipelines, and evolve their fundraising technology stacks. Of course, in each of these engagement, I’m always asked: “Where should we start with AI?” The answer, more often than not, is: start by becoming the kind of organization that AI can meaningfully support and improve.

Readiness is not just about installing more tools. It’s about aligning your people, data, and processes to unlock actual intelligence—not just artificial demonstrations of it. That’s what AI pragmatism is: the refusal to mortgage the future on hype, and the discipline to build the scaffolding for transformation that builds trust. Don’t get me wrong, I truly believe AI belongs in advancement, but only when it’s in service of real progress.