Intentional AI Use Starts With Asking Who It Actually Serves
What does it mean to use AI intentionally instead of just because it’s available?
It means weighing what a task actually needs against the real resources behind the tool, the electricity and water running through the data centers behind every prompt, before defaulting to AI. For nonprofit and small business leaders, that looks like matching AI use to an actual capacity gap instead of reaching for it out of convenience.
A local community group I follow online had a small fight recently. Someone suggested the moderators stop allowing AI generated product photos in the group’s marketplace posts. Within an hour the replies split into two camps. One camp said AI images already flood every feed, so singling out one small seller felt unfair. The other pointed out that a single mom running yoga classes had used AI to design flyers she never could have paid a designer for. Both were right, and that is exactly what makes this hard. Most conversations about “responsible AI use” stop at tone and disclosure. Very few name what is actually running behind the screen, or ask the harder question underneath it all: responsible to whom?
Why Responsible AI Conversations Skip the Data Centers
At Triple Creeks Consulting we help founder led nonprofits and small businesses build sustainable operations, and over the past couple of years that has come to include ethical AI adoption alongside our strategy, process and financial work.
We are not anti AI, and we use it in our own workflows. With our work implementing AI internally along with our work with clients, we keep noticing the same concerns. People feel uneasy about AI’s environmental footprint without having much to go on. A 2025 poll from the AP-NORC Center and the Energy Policy Institute at the University of Chicago found that 72 percent of Americans are at least somewhat concerned about AI’s environmental impact and 41% percent are very concerned. That is a lot of unease. Concern without information tends to curdle into guilt or denial, and neither one helps an executive director decide whether AI belongs in their orgs workflows.
What the Environmental Numbers Actually Show
The scale here is not small. Electricity demand from data centers surged 17 percent worldwide in 2025, and demand from AI focused facilities grew even faster. Training a single large model like GPT-3 evaporated an estimated 700,000 liters of freshwater in Microsoft’s US data centers, and total US data center water use has climbed sharply over the past decade as more facilities came online. None of this means every use of AI carries the same weight though. A short text reply and a generated video do not pull from the same pool. The International Energy Agency notes that newer AI applications like video generation and multi step agents can use hundreds or thousands of times more energy per query than a simple text prompt does. That distinction gets lost in most conversations about AI ethics, and it is one of the more useful ones available to us.
Responsible to Whom Is the Question Worth Sitting With
Here is the part that is harder to sit with. AI is not going anywhere because we wish it would. Platforms are already producing roughly 80 million images a day, and most of us scroll past dozens of them without a second thought; very few of us are reading anything about what those images actually cost to make. That gap, between how much AI content we see and how little we understand about it, is the bigger problem here, more than any single person’s choice to use a tool for a flyer. Every leader is already running their own cost benefit calculation, whether they realize it or not. A founder using AI to draft a grant proposal outline and a founder generating a meme for a group chat are not making the same trade. Neither one is wrong on its face. But naming the trade, out loud, is what turns “we use AI responsibly” from a slogan into an actual practice.
A Practical Way to Decide When AI Is Worth It
This is where we spend most of our time with clients, and it starts with a short set of questions before a tool gets folded into a workflow.
- What does the task actually need? A short internal email calls for far less computing power than a generated image or video, and knowing the difference changes which tool is worth using.
- Could a documented process handle this without AI at all? A good number of “we need AI for this” moments are really gaps in workflow or role clarity that a clear checklist or a defined handoff would solve just as well.
- Is this a one time use or a daily habit? Occasional use and an automated daily workflow draw very differently on the resources behind them, and the second one deserves more scrutiny.
- Are we choosing this because it serves the mission, or because it is the easiest button to press? That question alone stops a surprising number of defaults before they become habits.
We test this filter on our own operations before we bring it to a client, as part of the process development and operational structuring work we do. We do not recommend a system we have not used ourselves first.
What This Looks Like When You Use AI On Purpose
Using AI intentionally does not mean using it less out of guilt, and it does not mean using it constantly because it happens to be available. It means a team that can explain, in plain language, why a tool made sense for a specific task and what it replaced. It means asking who benefits and who pays before hitting generate. This is real responsibility.
If your organization is trying to figure out where AI actually fits and where it does not, that is exactly the kind of conversation we have with clients every week. Book a free discovery call and let’s map it out together.
