Teacher Grant Writing with AI, Pt. 1
TL;DR: Using AI as a Strategic Partner in Grant Proposal Development
(Instructional Guidance, Transcript, and Prompts below)
The Core Challenge in Grant Writing
The most critical factor in successful grant funding is not the final narrative, but the quality and alignment of the underlying project plan. A proposal must directly address funder priorities while meeting reviewer expectations. Failure to do either results in rejection, regardless of writing quality.
A structured review of the funding opportunity is essential. This includes identifying key priorities, areas of strong interest, and topics or approaches the funder is explicitly or implicitly not interested in. In this process, AI is most effective when used as an expert reviewer or grant-writing coach rather than as an automated proposal writer.
Process for AI Proposal Support
Start With a Concept Paper
The process begins with a concise concept paper that captures the project idea, need, context, and preliminary approach. This document serves as a working foundation and should be maintained separately so it can be revised iteratively.
Analyze the Concept for Alignment
The concept paper should be evaluated against the funding priorities to determine alignment, identify competitiveness risks, and decide whether the project is appropriate or salvageable for the opportunity.
Iterative Refinement Through Targeted Feedback
Key risks identified through review are systematically addressed by revising the concept paper. This may involve clarifying the program model, strengthening implementation details, or adjusting scope and emphasis.
Validate Revisions With a Second Review Pass
After revisions, the updated concept is reviewed again to assess how well weaknesses have been addressed and to identify any remaining concerns. Additional recommendations are incorporated as appropriate.
Strengthen the Implementation Model
When reviewer concerns remain abstract or difficult to articulate, AI can help translate recommendations into concrete implementation details and proposal-ready language that directly addresses those concerns.
The Outcome
The end result is a well-developed, reviewer-aligned project description and implementation plan. This foundation significantly increases the likelihood of funding success and provides strong material for the final proposal narrative.
9 Take-aways for Grant Writing
Below is a synthesized list of key takeaways, concepts, and guiding principles for teachers who want to write grants. These points reflect the underlying logic and method emphasized throughout the video.
- Funding decisions are driven by alignment, not effort.
If a proposal does not clearly address what the funder explicitly wants, it will not be funded, regardless of need, passion, or writing quality. A strong proposal emerges from a well-designed project model that satisfies funder priorities and reviewer expectations; writing quality cannot compensate for a weak plan. - Reviewer expectations matter as much as funder priorities.
Successful proposals simultaneously align with stated funding priorities and with how reviewers are trained to evaluate rigor, feasibility, and coherence. It is the reviewers who will score your proposal and determine its quality. - Understanding what funders are not interested in is critical.
Ignoring implicit or explicit exclusions is a common and avoidable reason proposals fail. - A concept paper is the foundation of proposal development.
Early articulation of ideas, context, need, and approach enables systematic refinement before committing to full proposal writing. - Concept papers should be treated as living documents.
Effective proposal development requires iterative revision, expansion, and refinement based on structured feedback. Repeated cycles of review and revision strengthen conceptual clarity, implementation detail, and methodological coherence. - AI is most effective as an expert reviewer, not a ghostwriter.
Using AI as a thought partner or expert coach produces stronger outcomes than delegating authorship to AI. - Risk identification improves competitiveness.
Explicitly identifying weaknesses, misalignment, or ambiguity allows proposers to proactively address reviewer concerns. - Sample language accelerates refinement but requires human judgment.
Proposed language should be evaluated, adapted, and intentionally integrated rather than copied wholesale. - Proposal development is a strategic design process.
The goal is not simply to “write a grant,” but to engineer a project that fits a specific funding opportunity.
Full Transcript and Prompts
Most Important Part of Grant Writing
So, you’re going to write a grant.
You have a special project or a special plan or special need and you need some funding for it. So, you’re going to write a grant.
I will tell you, having written a lot of grants over the years, the most important part of the process—and the most challenging part of the process—is developing a proposal plan, a project plan that’s going to address the funder’s interest and meet the reviewer’s expectations. If you can’t get those two parts right, you’re not getting any money.
Fortunately, we can turn to AI for help with that. And that’s what we’re going to do here.
Develop an Initial Concept Paper
The very first thing that we’re going to do is we’re going to create a concept paper. Basically, what are you thinking about so far? What are your ideas? What are your needs? What is the context? What’s it about? What have you already got here? and you’re going to write it down into a concept paper.
Now, you could take this information and just put it into the chat box, you know, just write it in there. I recommend you do it on a separate document and upload it because we’re going to come back and do some revisions to it. We’re going to take some stuff out and add some stuff to it and so forth. So, we’ve got that.
We’re going to upload that concept paper to our AI, and we’re also going to upload the Request for Proposals, the document that says “Here’s the opportunity, and here’s what it’s about, and so forth.” So, we’re going to upload those, and we’re going to ask the AI to do a few things for us.
RFP Concept Alignment
Prompt
First Review the uploaded Call for Proposals (RFP), with a focus on
(a) key priorities,
(b) areas of strong interest, and
(c) areas that [the funder] is explicitly or implicitly not interested in
Second, describe alignment between the RFP and my project concept paper. Identify any risks that might hurt proposal competitiveness
Third, based on your review, do you think my concept is appropriate for this funding opportunity?
First, fully review and analyze the request for proposal. What are the key priorities? Like, what’s the funding for, right? What are the areas of interest by the funder? And, very importantly, which we often forget, what are some things the funder is explicitly not interested in?
Our proposal, our plan, needs to address the things they want, and it needs to avoid the things that they don’t want. Otherwise, we can kiss the money goodbye. We’re not getting it. So, that has to be done right. That’s the first part.
Then with that now take a look at the concept paper. Take a look at the RFP and do an alignment. What is the alignment?
And, importantly, identify any risks that might hurt the proposal competitiveness. It’s those risk that are going to tell us what do we need to do next.
And, finally, having done those two analyses, are we on the right track? Do we have a chance? Is it salvageable? Can we make it work?
Based on your review, do you think the concept is appropriate for this funding opportunity?
We’re going to turn that prompt over with the concept paper and the request for proposals to AI, and it’s going to tell us, basically, what do we do now?
So, interest, non-interests here. (We don’t need to worry about what this content is. It’s very specific to this RFP. It’s the process that’s the important thing and the type of information that the AI is providing. All right?) Alignment in this particular case, it’s no good. Finally, it’s not yet well suited but it is salvageable. Okay.
Now that we know it’s worth doing the work, let’s figure out what the work actually is. So, your concept would need to…and overall here’s what we’re looking for, and it’s all in there.
So that brings us to the next prompt.
Detailed AI Concept Review and Revision
Prompt
You are an expert reviewer and grant writer for this funder.
With the proposed ideas regarding the project details, what are risks to successful submission and reviewer approval?
How might the key risks or misalignment be addressed?
The next thing we’re going to do is say, “Okay, you AI, you are an expert reviewer and grant writer for this particular funder.”
See, in this case, we’re not saying, “Okay, AI, I need this grant. Go off and write it for me.” You’re not getting any money if you do that. What we’re doing instead is we’re having the AI serve as an expert team member, or an expert level coach, who understands this process and can help guide us through preparing an effective and successful proposal. Right?
We’re not turning the work over. We’re using the AI as a “thought partner” to help us to develop a strong proposal. There’s still work to do. We still have work to do. The AI is going to help us decide what the work is and how we can go about it.
All right. So, you’re an expert thing, an expert reviewer, regarding the details. What are the risks to successful submission? If we turn it in like this, what are the risks that we’re not going to get funding.
And here’s the important bit. (It’s all important.) How might the key risks or misalignments be addressed? What are the problems? What do we do about them?
And it’s [AI is] going to give us some very solid feedback. I will tell you, these AI models have gotten pretty good at this.
So, here are big problems, risks, why it hurts the competitiveness, and what can we do about it, including some sample language. Now, here’s what we’re going to do with this. We’re going to go back to our concept paper with this in mind after we think about it with our team.
What do we need to take out of our concept paper? What do we need to add in? What details do we need to provide? And, as you see, there’s even some sample language if we agree with the recommendation. Here’s some language we can add to make it work.
This, then, all gets fed back into our concept paper for a better description of what the project is. And, as you can see, now that I’ve done that, it’s gotten quite a bit longer, which is fine. That’s what we needed.
Second AI Pass for Specific Proposal Content
Now, following that, we’ve updated our concept paper, and we’re going to feed it back into the AI. And that’s why we did it on a separate document so we can make some changes and feed it back in.
Prompt
Review this updated concept paper. To what degree do the revised ideas address
weaknesses, and what recommendations do you have to make the project model better?
So, review this updated concept paper. To what degree do the changes, the revised ideas address the weaknesses? And what recommendations do you have to make it even better? And it will come up with some.
This seems like a lot of back and forth work, but remember, we’re not turning over the task to AI. (Go do it!) We’re collaborating with the AI as a team member here.
Okay. So we still have some problems. Let me zip down to where it is because that response went on for a while. (Good. Good stuff. Okay. Finally, there it is. All right.) Degree to which it addresses the problem. So, great! What improved and are remaining concerns.
Now, check this out: recommendations. If that’s still the remaining concern, here’s what you want to do about it to fix the problem.
Basically, we review them [the recommendations]. We say, “You know what? I can do that. I agree. I understand. It’s something I can incorporate. I can approve of it.” That is going to go into my concept paper.
And it’s going to keep going and keep going and keep going.
It did a pretty thorough analysis, quite honestly. Look, even session plans that we might use. Lots of good stuff.
Assistance Addressing Possible Concerns
Prompt
What might this look like in regards to the possible program implementation model and reviewer concerns?
Basically, with this approach, propose language to address reviewer concerns.
Okay, so finally it’s done. It’s come up with some suggestions, and very likely it says, “Hey, think about this and include that.” And we come back and say, “I’m not really sure what that would look like. How would I go about that? How might I state that within my project in my proposal?” So we ask.
AI says, “Here’s an idea.” Well, tell me more. Okay, so we copy one or more of the recommendations. I would do just one at a time. How might this look like in regards to the program implementation and reviewer’s concerns? And, basically, propose language to address those reviewers’ concerns.
And, as you’ll see, it’s going to give us some pretty good (if I can ever get to it), ideas what it looks like.
If we zip past here, we go concerns. And this is language that we can drop straight into our proposal.
Finalizing the Project Description
Now, what we’re going to do with it here is we’re going to drop it into our concept paper because it all becomes part of the project description. So, at some point, we’re going to get the AI to help us put it all together and actually write this thing, and it’s going to draw from this. So, we’re going to actually feed it some valuable and productive language.
At the end of the day, what we should have is a very strong program model or implementation plan because that, more than anything else, is what helps us to get the funding.
I hope you found this to be useful. Take care.