You’re no doubt aware that generative AI has become a hot topic over the past few years. Whether you love it or hate it, grant writers have options: run for the hills or grab hold and go for the ride.
What’s the Big Idea?
Generative AI is here and it’s advancing—fast. I won’t get into the ethical debates around using it (think ChatGPT, Gemini, Claude, Copilot and all the ways they were initially trained, not to mention the environmental impact); there are plenty of online spaces where people can share information and discuss these topics.
My focus is on the output. Generative AI (let’s just call it AI here) can be a valuable tool in the grant writer’s toolbox. Users should, however, be wary of its shortcomings.
AI in Action
First off, AI responses are only as good as their prompts. Without comprehensive prompts, you’re unlikely to get valuable content. (Check out this post from MIT on creating good prompts.)
Once you develop this basic skill, AI can be a great starting point to shape your application. It can help to quickly determine your eligibility and review funder priorities. It may even be able to connect these priorities with your specific project/program goals.
AI can help to outline your application responses. You can use it to brainstorm ideas (big picture budget items, evaluation methods) or ask about potential frameworks (SMART goals, logic models, theories of change). It can pull keywords that may be valuable to use (think of those funder priorities) or cut text to match word/character restrictions. But—BUT—be ready to critically assess the content produced to ensure it’ll suit your objectives and resonate with the funder.
Does AI Have Limitations?
Do otters have pockets? (The answer is yes!) The material that AI presents is often generic and repetitive. AI lacks contextual judgement, making its responses bland and generalized—something grant reviewers can easily spot.
Furthermore, any “evidence” (statistics, citations, etc)—if given at all—will need to be vetted and confirmed. The rate of hallucination (fictional information) is still high and AI wants to please us. Its responses will more often than not seem like they’re just what we need. But it tends to fabricate details like hard data, success stories, and sources. Without the right prompt it’ll make endless assumptions, which are often incorrect. With the time it takes to research and re-write these sections, it may be more efficient to draft from scratch with all the relevant information at hand.
⚠️ I advise people to be cautious in what they share through AI prompts and uploads. Unless you access a paid plan with strict security and privacy settings, any information you share is most likely used to train the algorithms. You may also want to check that you’re not at risk of breaching privacy or confidentiality policies when using AI tools.
Knowledge is Power
Humans remain more powerful (at least, at the time of writing). Successful applications aren’t full of generic content. They have narratives, developed by thoughtful writers who can create a connection between the audience and project/program goals, while seamlessly weaving in funder priorities and expectations. Good grant writers provide strategic insight and understand nuance. They can anticipate the direction of the project/program and demonstrate its value to reviewers with diverse perspectives.
Don’t rely on AI to secure funding. Tap into human capacity—invest the time to make your proposal stand out by hiring a grant writer or editor (👋).
🔜 Next time: we’ll get into visual formatting—an often-overlooked element that has a major impact.
Thanks for reading! I’d love to know what resonated with you or any specific components you’d like to learn more about. Leave a comment, use the Contact page, or email me directly. Until next time!
jbr editing 