Quick Tips
What Empowerment Language Hides in Your Narrative
I recently read a post on LinkedIn about “empowerment” language, and it really got me thinking! Language is a living, breathing part of our work, and it changes constantly. When I first started grant writing more than a decade ago (now closer to two decades than one!), “empowerment” language was just entering grant writing, and it was considered fairly progressive. The idea at the time was that we were moving away from the nonprofit being the one doing the work to make the change to providing support to a community so they could make the change.
More recent critiques of empowerment language argue that it often overlooks what is already present in communities and still gives the organization disproportionate credit for the work that happens.
There is a sentence that lives in nearly every grant proposal I have ever read, and in plenty I have written myself: "Our organization empowers [women] to [build financial independence]." Insert your variable between the brackets. It reads as warm, mission-driven. It has done its job in thousands of funded applications. I want to slow down on it, because that sentence carries an assumption about who matters most in the work.
Empowerment language has become the sector's default vocabulary. It signals values. It tells a funder you care. It also, in the same breath, places your organization at the center of the story and the community you serve at the receiving end of your generosity. That arrangement is worth examining, especially for those of us who care about equity and want our grant narratives to reflect it.
Why AI Grant Writing Tools Fall Short, and What Builds Real Capacity Instead
Every few months, a new tool promises to write your grants for you. Plug in your organization, answer a few prompts, and out comes a proposal. For a nonprofit leader trying to expand a grants portfolio with a lean team, that promise lands somewhere between exciting and too good to be true.
It is usually the second one. The tools are getting better, and some of them are genuinely useful for a first draft. Here’s the problem with these tools: when you try to generalize something for everyone (selling a product to 1000s of nonprofits), the specifics get lost. And in grant writing, the specifics are the entire point.
Why Generic AI Grant Writing Tools Lose What Matters
A general-purpose AI grant tool is built to serve thousands of organizations at once. To do that, it has to flatten the way that it writes grants. When a tool is designed to work for everyone, it cannot hold the things that make your application yours. You get language that is technically competent and fairly interchangeable. A reviewer reading a hundred applications can feel that flatness immediately.
The deeper issue is what happens when the output is wrong. With a closed tool where you can’t rewrite the skills/code, you are stuck. You can regenerate; you can tweak your inputs, but you cannot reach into the system and fix what is actually broken because you did not build it and you do not understand how it works.
The Case for Training Your Team Instead
There is another path: when you train the experts in your organization to use AI effectively, they can fix the problems themselves.
I have been saying this for a long time (in AI years…so like at least 18 months…ha!): the real promise of AI is that you get to solve YOUR problems. The accessibility of the tech is the benefit.
AI Certifications in Grant Applications: What This Really Means for Grant Professionals
A client recently asked me to sign a certification stating that no part of their grant application involved AI. As soon as I read it, my stomach tightened.
My unease comes from knowing that signing that certification with confidence is a lot more complicated than the form makes it look and will become increasingly challenging to promise.
These certifications deserve more conversation than they're getting. The intent behind them is reasonable. The implementation is creating problems that funders may not fully understand.
Let’s level set:
Funder AI certifications have grown more common in the last year; I’d never seen one before 2026.
The tools to enforce this are unreliable.
False positives in AI detection tools are widespread and well-documented. As I read recently on LinkedIn, someone’s essay on " The Yellow Wallpaper” from high school English class back in the ‘90s gets flagged as AI generated. So do my standard grant narratives pre-2023.
Grant writers may unknowingly sign inaccurate certifications when clients use AI upstream. (Think: that really nice program description the program manager sent you or the annual report that you pulled language from? Could absolutely, and increasingly, so, be generated by AI).
The field lacks a shared definition of "AI use," making honest answers difficult. Are editing tools with AI built in “AI use”? What about using AI to order your citations? Or review your draft and help you improve it? Or create your logic model from your narrative?
Shadow Work in Grants
Shadow work is the hidden labor behind every successful grant—from rebuilding budgets to chasing internal data. New research shows it’s costing organizations an average of $53,700 a year and forcing many to walk away from major funding opportunities. This post explores what’s really happening behind the scenes and why it’s time to rethink how grant work is supported.