top of page

Worth The Experiment? - Does AI Actually Save a Tiny Arts Organization Time?

Writer: Tanesha Ford
Tanesha Ford
Nov 8
7 min read
Mid-century modern illustration of an empty, angular captain’s chair on a retro-futuristic spaceship bridge, surrounded by geometric control panels and a panoramic view of colorful planets, stars, and a ringed celestial body. Rich mustard yellow, moss green, burnt orange, walnut brown, and deep blue create a vintage space-age aesthetic. Text reads “Marketing Moments,” with the For de Arts Consulting logo in the lower right corner.
This 11:22-minute audio sound bite has been digitally created to provide an accessible listening option.

What happens when we measure the whole process, improve our skills, and consider who gets access to the tools?


I grew up watching the original Star Trek with Captain James T. Kirk alongside my mother, sitting on our brown velour couch in Fort Dix, New Jersey. Through her, I developed a deep and abiding love for science fiction and fantasy. As I got older, Captain Jean-Luc Picard became the captain of MY heart, and I happily followed him and his motley crew throughout the galaxy.


Following the passing of the beloved Patrick Stewart, I found myself revisiting Star Trek: The Next Generation. As it so often happens, when I am binge-watching anything, my mind began wandering toward the parallels between the Federation and the nonprofit arts sector.


I started thinking about the ship's computer and the people responsible for deciding what to do with all that information. How might that relationship translate to our everyday work, especially within arts organizations operating with limited resources?


So, pour yourself a cup of Earl Grey tea, hot, find yourself a secure seat, and let's engage.

The Computer Is Ready. Are We?


Artificial intelligence has become a rather unavoidable topic in nonprofit circles, and the conversation is frequently dominated by two competing narratives. One promises extraordinary productivity, while the other raises legitimate concerns about creative ownership, employment, environmental costs, and the consequences of allowing technology to make decisions we could be making ourselves.


Both deserve our attention.


According to Google's 2025 survey of more than 9,000 nonprofit organizations, 68% of respondents reported that at least some members of their organizations were using generative AI, while 73% of respondents using Google's AI tools reported spending less time on administrative tasks.


Meanwhile, Salesforce's Seventh Edition Nonprofit Trends Report found that more than half of surveyed nonprofits were piloting or using AI in their daily work, even as privacy and data security remained significant concerns.


Those numbers suggest that organizations are experimenting, and many are finding value.

However, a reported improvement in productivity tells us relatively little about what happens between generating an impressive first draft and delivering something worthy of an organization's name.


For a small arts organization, that distinction matters enormously.

Let's Put the Computer to Work


Imagine a small theater company preparing its monthly newsletter. Its administrator is also managing ticket sales, coordinating volunteers, answering patron questions, and trying to remember whether anyone confirmed the photographer for Saturday's performance.


The newsletter still needs to go out.


Without AI, the administrator gathers program information, writes the copy, edits it, checks dates and links, formats the email, and schedules distribution.


With AI, perhaps the first draft appears in two minutes. Wonderful! But the administrator must still review the language, correct inaccuracies, restore the organization's personality, verify the details, and make sure the final message serves its audience.


Did AI save time?

Possibly. We haven't finished measuring.


And here's where I want to introduce another consideration: anything new takes practice.


If your first attempt at creating an AI-assisted newsletter takes 90 minutes, that tells you how long your first attempt took. It tells you very little about how efficiently you'll perform the same task after five, ten, or twenty attempts.


Think about how long it took you to learn your email marketing platform, navigate your organization's database, or build your first campaign report. Familiarity changes the equation.


AI is no exception.

Training the Computer Changes the Equation


One of the most useful things you can do before asking AI to write anything is to teach it about your organization.


Its personality, tone, audience, mission, and communication preferences can all be established upfront. You can provide examples of previously approved newsletters, explain the language your organization uses, and identify expressions or approaches that don't belong in your communications.


Some AI tools allow you to save these instructions for repeated use, while others require you to supply that context again.


Either way, you're giving the technology a better understanding of what you're asking it to accomplish.


Instead of spending your review process transforming generic language into something recognizable, you're starting with material that should already be closer to your organization's voice.


And that brings us to another opportunity: improving how we check the work.


Imagine you've supplied an AI tool with the original articles, program descriptions, and event details needed to create your newsletter blurbs.


After generating the content, you review the first draft, correcting obvious inaccuracies and adjusting language where necessary.


Next, download a test version of your newsletter and upload it back into the AI tool alongside the source material. Ask it to compare the finished newsletter directly against those sources, identifying discrepancies in dates, names, descriptions, links, and other factual details.


This second review can catch inconsistencies you may have overlooked, particularly when you're working with multiple stories and deadlines.


And when something in the content doesn't align with what you believe to be true, ask AI to cite its sources. Better yet, ask it to identify the exact passage in the original material supporting the statement.


Then examine that source yourself.

AI can generate convincing citations that are inaccurate or nonexistent, and an automated comparison can miss mistakes. These methods help focus your attention, but they don't eliminate the need for human verification.


Over time, you can refine your instructions based on the mistakes that keep appearing, making future drafts more accurate and easier to review.


The first newsletter may take 90 minutes. The fifth might take 60. The tenth might take 35.


Those are hypothetical numbers, of course, but they illustrate why evaluating AI solely on your first attempt can produce a misleading conclusion.


The more useful question is whether your investment in learning the technology produces meaningful improvements over time.

What Are We Actually Trying to Improve?


During my nearly 15 years working in the nonprofit arts sector and 15 more in the for-profit sector, I've come to know that capacity challenges rarely present themselves as a tidy list of tasks waiting to be completed.


They look like 45 things that need attention, three people available to handle them, and the understanding that some of the work will receive less care than it deserves.


What interests me about AI is the possibility of changing that equation.


Suppose an organization currently manages 45 responsibilities with varying degrees of success. Could thoughtful use of AI help its staff concentrate on 30 priorities and execute them with greater consistency, creativity, and care?


Perhaps AI helps organize audience feedback, prepare a first-pass content calendar, or identify inconsistencies in campaign messaging. Those contributions could create space for staff to strengthen relationships, improve accessibility, or develop programming that genuinely responds to their communities.


The real measure of success would be the value of that reclaimed attention.

The Enterprise's computer could process enormous amounts of information, but Captain Picard still had to understand the circumstances, consult his crew, and take responsibility for the decisions.


Our organizations deserve the same discernment.

The Equity Question We Cannot Ignore


Here's where the conversation gets more complicated, because the opportunities and consequences of AI aren't experienced equally across the arts sector.


There are organizations that have never hired a graphic designer, copywriter, marketing consultant, or communications strategist because those services have always been financially out of reach.


Some lack the staff capacity to pursue grants that might eventually provide access to those resources.


When we discuss AI primarily through the lens of replacing creative labor, we risk overlooking organizations that have never had the means to purchase that labor in the first place. I want us to hold both realities in the same conversation.


Artists deserve protection from unauthorized exploitation of their work. Creative professionals deserve fair compensation. Organizations have a responsibility to understand the legal, ethical, and privacy implications of the tools they use.


And small, under-resourced arts organizations deserve meaningful access to tools that may help them communicate, organize, and participate more fully in their communities.


These responsibilities can coexist.

The question becomes how we establish thoughtful boundaries while expanding opportunity. That includes knowing when to pay an artist, when AI can responsibly assist with routine work, and when the technology simply isn't appropriate.


Access without discernment creates another set of problems.

Discernment without access can preserve existing inequities.

Before We Beam Down, Check the Coordinates


Perhaps the most revealing question an arts organization can ask about AI concerns its own intentions.


Are we exploring this technology because we want to eliminate expenses previously devoted to creative professionals? Are we accepting mediocre results because they arrive quickly? Are we evaluating the output with enough knowledge to recognize when something is inaccurate, derivative, insensitive, or simply ineffective?


Or are we looking for a responsible way to accomplish work that has exceeded our capacity?


These motivations deserve examination because they influence the decisions that follow.

AI can produce confident language without understanding the community an organization serves. It can suggest strategies without knowing the relationships that make those strategies possible. It can generate images and ideas while raising unresolved questions about consent and ownership.


Human judgment remains essential, and organizations need to account for the labor required to exercise it.

Engage, With Intention


The promise of AI deserves investigation, especially for organizations accustomed to stretching limited resources across enormous expectations.


The risks deserve equal scrutiny, and so does the learning process.


I'd encourage any small arts organization considering AI to select one recurring task, document its current process, and conduct a limited experiment. Measure the entire workflow, evaluate the quality of the result, and identify where human expertise remains indispensable.


Then repeat the experiment.


Improve your instructions, establish your organization's voice, develop a consistent method for checking sources, and measure the process again. Give yourself the opportunity to become proficient before deciding whether the technology has earned a place in your workflow.


And ask what the organization would actually do with the time recovered.


Would it strengthen a relationship with an artist? Improve the patron experience? Give an overextended staff member room to think? Make space for work that has been postponed for months?


Those outcomes may tell us far more than the speed of a generated response.


The Federation's technology was extraordinary, but its aspirations were rooted in exploration, cooperation, and the belief that knowledge could serve a greater purpose.

That's a useful standard for our sector, too.


What recurring task would be worth testing if you measured the entire workflow, including how that workflow improves with practice?

Research referenced


Comments


For de Arts

Marketing Moments •

Thoughtful, practical conversations about nonprofit arts marketing, audiences, capacity, and the things small teams are actually navigating.

‪(609) 534-0298‬

​

  • LinkedIn
  • Instagram
bottom of page