During the summer of 1983, my sister and I walked out of WarGames fascinated by two things. We both had a crush on Matthew Broderick. And we were both mesmerized and a little bit scared of Joshua – the computer that learned.

A Brief History

Nearly twenty-five years later, an IBM supercomputer, Deep Blue, beat then-world chess champion, Garry Kasparov, in a 1997 chess tournament in New York City. It was the first time a reigning world champion was beaten by a computer and the first time the public could really see the power of Artificial Intelligence (AI).

Twenty years after Deep Blue, another computer—AlphaGo—defeated world champion, Lee Sedol, in the game of Go. This defeat was even more significant.

And that’s because the game of Go is much more complex than chess—it has a larger board, longer games, and infinitely more moves to consider on each turn.

AlphaGo used neural networks and an algorithm that uses random simulations to estimate its probability of winning. It also accessed and analyzed the entire online library of Go, including all matches, players, analytics, and literature. After its initial development, AlphaGo was no longer dependent on programmers for improving the algorithm and search function—it could do that on its own. . .and learn from past games to improve its chances in the next game.

Today

Since then, AI-powered tools, platforms, and applications have dominated the landscape. AI is behind the personalized recommendations you get on shopping sites like Amazon and entertainment platforms like Netflix. AI is what makes facial recognition work, sets your house temperature to the optimal degree, and recognizes your speech when you are dictating an email.

Even before AI, tools like ChatGPT had us testing if we could distinguish it from a human writer. AI was entrenched in our everyday lives. And it was also embedded in our nonprofit management, fundraising, and marketing programs.

Artificial Intelligence in Today’s Fundraising Programs

The question we face today is not if we’re going to use Artificial Intelligence, but where and how.

From writing more compelling subject lines to finding new donors, AI platforms and tools can be applied to all areas of fundraising. Employing AI makes everything, from ads to analytics to content, more intelligent.

AI can analyze donor and prospect data, predict which donors are most likely to give a gift, recommend the most fundraising actions to take, forecast results, optimize ask amounts, copy, design, and much, much more.

Here are six ways that AI can help improve your fundraising programs (and aren’t that expensive or difficult to implement):

1. Donor segmentation: AI can analyze donor data to identify patterns and segment donors for targeted fundraising campaigns. Translation: AI tools can tell you which donors to solicit in your upcoming appeal and how to group them.

What really makes an AI-powered donor segmentation tool impressive is the ability of the machine learning to improve over time. Instead of a fundraiser or marketer giving rules for who to select and who not to, the tool will continually learn based on new inputs, and the segmentation will get more targeted and more productive over each outreach campaign.

The more first-party data you provide (gift date and gift amount, but also channel, gift type, themes, signers, tone, design approach, outreach strategy, tactics), the better prediction AI will make for your targeted audience. This is why having consistent and robust nomenclature, attributes, and tags for your outbound communications is so critical.

2. Personalized communication: AI can help tailor communication to individual donors based on their giving history and preferences. This includes the topic, tone, signer, and graphic approach.

Using natural language processing (NLP), an AI-powered NLP tool can analyze your donor interactions and feedback to understand their preferences and needs. This information can target donors with relevant and personalized messaging and outreach targeted to their interests, preferences, and behavior. For example, after reviewing all of your subject lines and engagement, AI can generate new, emotion-based subject lines that increase open and click-through rates.

Other AI tools optimize design for maximum impact, such as by selecting images that are most likely to catch the donor’s attention or choosing colors that are most appealing to the donor segment.

3. Predictive analytics: AI can help predict future donations and future fundraising campaigns. At a high level, there are programs that can continuously evaluate your fundraising outreach and tell you how future campaigns will perform. While fundraisers do this now, augmenting your human intelligence with AI allows you to analyze more data points and make connections within the data that may be missed.

At an individual donor level, AI can determine not just when you should ask and what you should ask, but how much you should ask for. There are several tools that estimate the donor gift amount and build customized ask strings to accompany this targeted ask. Some tools will also predict if you should be asking for a one-time or sustainer gift.

4. Chatbots: AI-powered chatbots can provide quick and personalized responses to donor inquiries and help manage donations. Chatbots can increase website conversion rates and decrease donation page abandonment rates by interacting with donors in real time.

One study showed that 62% of consumers would use a chatbot to help them with their purchase instead of waiting for a customer service representative. 90% of people used a chatbot within the last year and 70% of chatbot users had a good experience.

 When it comes to chatbots, basic is often better. This is great news for nonprofits—you don’t need to be a huge business or organization to implement and benefit from having a chatbot. A simple rule-based chatbot, which relies on a set of predefined rules and is designed to handle specific tasks or answer specific questions, is straightforward to implement and can help with basic donor services requests.

5. Image and voice recognition: AI is being used to automate tasks such as identifying and categorizing images and videos, and transcribing speech to text. AI can also recognize images of donors and their voices, which is being touted as the next way to create a personalized experience and increase their engagement.

Of course, voice recognition can also be used to donate. For donors that are comfortable using voice commands, organizations can ask donors to update their payment information, make a gift, and confirm their address, and the donor will respond via voice.

6. Optimizing fundraising strategies: AI can analyze data from past fundraising campaigns and suggest strategies for improvement. This is a machine learning type of AI, which identifies patterns based on large sets of data. The machine uses these patterns to make predictions. Then it uses more and more data to improve those predictions over time.

AI-powered tools can also optimize your postal and email prospecting lists. There are many databases that use AI to help fundraisers select the most productive names for lead and donor acquisition.

These are just a few areas where you can use AI to improve your fundraising program. There are many more ways that AI is already being used in your campaigns and even more that are still to come. As AI tools and platforms become more sophisticated —and more ubiquitous—nearly all data-driven tasks will use some form of AI.

But AI is not without its complications. There are serious ethical questions, including bias and discrimination, privacy, lack of accountability, and transparency. These are all being debated now. As fundraisers, we must have these conversations and understand how AI is being used and how it was built.

Of course, it’s just as important to know where AI is beneficial to your organization and where
it isn’t.

Proceed with Caution

Copywriting: Recently natural language AI or natural language processing (NLT) tools and platforms have been in the news, which has prompted many people to test them for writing direct mail, proposals, emails, etc. Tools like ChatGPT, a conversational AI model developed by OpenAI, generates human-like text from a user prompt. But, even according to ChatGPT itself, “ChatGPT is not a perfect system, and its responses may
not always be accurate, appropriate, or free of biases.”

It also isn’t very warm. The conversational tone of ChatGPT is very different from a fundraiser’s voice. And it can only generate responses on what it has access to. New information that’s not on the internet won’t be included in its answers.

As one copywriter wrote, “It gives you a first draft. You’re still miles from a finished piece of fundraising copy. You have a lot of work to do to create connecting, human, effective writing.”

So true. This entire article was written via ChatGPT—and took hours of editing, rewriting, and fact-checking.

One day, an AI tool will produce effective outreach copy. But we’re not quite there yet.

Copywriting aside (for now), the AI revolution is here —and has been for a while. Every day brings new AI-powered tools for fundraisers to investigate and adopt. Fundraising isn’t easy, but AI products can help save hours of time and improve your campaigns. Just know what you’re using and how much you can trust the tool.

(Originally posted on Nonprofit Times)