AI Adoption Is a Training Problem, Not Just a Tool Problem
Most organizations already use AI. Far fewer get real value from it. The gap is training, habits and clear rules.
Buying an AI tool is the easy part. Getting a team to use it well, safely and consistently is where most organizations stall. The numbers make the point: a 2026 survey of nonprofits found that 92% use AI in some capacity, but only 7% report major improvements in their ability to deliver on their mission. Nearly half of the organizations not yet using it cited a lack of training, only a handful had an AI training budget, and nearly half had no AI policy at all (Virtuous, 2026 Nonprofit AI Adoption Report).
Why tools alone don't change work
When people get access to an AI tool without guidance, three things happen. A few enthusiasts use it constantly and inconsistently. Most people try it once, get a mediocre result and stop. And someone pastes something into it that should never have left the building. None of these are tool problems.
What effective AI training looks like
- Built on real tasks. Train on the emails, reports and documents people actually produce, not generic examples.
- Practice, not presentation. Every person completes one real task during the session.
- Clear "never paste" rules. Client data, donor and health information, passwords and anything confidential.
- How to check outputs. Verifying facts, spotting confident errors, and knowing when to start over.
- Reusable patterns. Prompt cards for each role, so good results don't depend on memory.
- A follow-up. A check-in after about 30 days to see what stuck and fix what didn't.
Write a one-page AI policy
A policy doesn't need to be long to be useful. One page can cover what tools are approved, what information must never be entered, when AI-assisted work must be disclosed, who reviews outputs before they go out, and who to ask when unsure. The point is to give people confidence to use AI, not reasons to avoid it.
Measure adoption like any other program
Track how many people use the approved tools weekly, which tasks they use them for, and what they say it saves. Identify a few internal champions who can answer questions. Adoption is a program to manage, not a switch to flip.
The organizations getting value from AI aren't the ones with the most tools. They're the ones whose people know what to do with them.