AI and grants fundraising

Are you using AI to support your grant fundraising activities yet? Based on the conversations we’ve been having with funders about our Funder Plus work recently, if you’re not using it, you’re very much in the minority. One funder we spoke to said the majority of the 700 applications they received for a recent funding round showed at least some signs of having been put together using AI.

Oddly put together sentences, strangely placed superlatives, American spellings and sweeping generalisations are the usual giveaways – although the funders that have made statements on the use of AI seem to be less concerned about its use as they are about it actually improving the quality of the applications they receive (NCLF guidance on AI).

The good news for now is that most funders are a little behind the curve when it comes to AI. You might have these amazing new tools at your disposal as an applicant, but the questions, forms, assessment and shortlisting processes, due diligence etc. hasn’t changed (yet*). Large funders using AI to sift and manage the huge numbers of applications they get probably isn’t that far away, but for now your goal is the same as it’s always been – communicate the need for your work, show how it matches the funders priorities, and demonstrate the impact it will have, better than everyone else who is applying for the same funds.

So how can you use AI to make that easier, and improve your chances of success?

1. Improving your stock copy: Collating and repackaging information is one of the most time-consuming tasks that AI can manage easily. If you have a single comprehensive document, set of reports, or just your website, it’s easy to get ChatGPT or something similar to produce endless variations of it, tweaked to highlight specific priorities.
2. Tell better stories with your data: Sorting and matching data can also be a doddle with AI. Tools like ai can help to build visuals and give you great insights into what your data is telling you, and help to underpin the stories and case studies in your applications.
3. Know more about who you’re applying to: Want to get a bit more insight into how to tailor a letter of application? Try asking ChatGPT “What other charities are the trustees of [this trust] involved with, either as staff, volunteers or trustees?” and get a near instant list of the other causes and organisations they are supporting.
4. Target copy quickly: Need ANOTHER response to the question “What impact does your organisation make?” but this time it has to be 650 words and you want to prioritise work that’s taken place in the North West with children and young people – easy peasy, upload all the slightly different answers you gave the last fifteen times, ask for what you need this time, and you just saved yourself an afternoon of screaming at word counts.

So the upsides are great, more time, quicker production of copy, no more writers block or battling word counts, what could go wrong?

Well, plenty if you’re not careful.

It’s really easy to lose any sense of authenticity when you’re squeezing your copy through ChatGPT or perplexity et al. They can filter, sift, construct sentences with emphasis on key words, but they can only work with what they’ve got and what you give them. It’s easy to end up with generalised copy that doesn’t feel grounded in anything meaningful. Artificial intelligence is one thing, but artificial empathy often reads horribly – and don’t forget, at least for now, it’s a human being that’s reading your application (alongside hundreds of others) looking for the ones that connect best with what they are trying to achieve as a funder.

You can work on this to some extent by developing really clear prompts, give examples of existing copy and ask for new copy to match the style or use specific words, but you’re still unlikely to get a perfect and useable response that doesn’t need any editing.

There are also data and privacy issues with a lot of (especially free) AI tools. With some paid for tools like Copilot, any data or details relating to your work that you feed into it will be secure and safe. With others, anything you upload or paste into it will be stored and used to further develop the platform in question so be careful with anything that’s not already in the public domain.

Then there are the moral and ethical issues. Generative AI can replicate the biases and paradigms of both its creators and the information or data that it’s drawing on to produce copy, cases studies, data summaries or other information you might want to include in a funding application. AI also has a huge and increasing environmental impact. An AI search uses ten times more energy than a traditional search engine, as well as the increased demand for components and the raw materials required to deliver the computing power required to deliver it.

Running through many of these issues is integrity – of the content you’re producing, the origin and impact of the tools you are using, and the way in which you are using it. Many charities (including DSC) are developing AI policies to manage its use within their organisations, and if you’re doing the same it’s certainly worth looking at where fundraising fits within any organisational approach. While there’s no objectively right or wrong way to use AI in your fundraising, having a considered and consistent approach might help to reinforce applications to certain funders. A simple sentence within an application stating something like “The content of this application was produced in part using generative AI in accordance with our organisational AI policy.” might reassure a grant assessor that any oddities or AI giveaways are part of a proactive approach rather than a clumsy attempt to shortcut their processes. On the other hand, saying something like “As part of our commitment to reducing our environmental impact and challenging structural biases, this application was not produced using any generative AI tools” in an application to an environmental funder may land well**.

In summary there are a lot of opportunities for using AI to support grants fundraising, but a lot of the tried and tested tips still hold true, like reading all of the available information from a funder before you apply (including whether they have a policy on using AI), and getting a human being that isn’t you to read your application before you send it off (whether you wrote it or ChatGPT did!

To learn more about AI and the impact it is having on charities, check out our upcoming course here

To keep in the loop on our work on AI and get more articles and info like this, sign up to our enewsletter here

*AI grant assessment is not that far away, check out this from Community Force in the US

**this article was written the old fashioned way, all typos, mistakes and generalisations were made by a human being