What Tribal Business News Got Right About AI in Tribal Operations — and the Part They Underplayed

Tribal Business News ran an article this month on how tribal governments and tribal enterprises are starting to use AI in practical, business-focused ways — accounting automation, invoice processing, grant writing, visitor analytics, customer segmentation, language preservation, and a handful of others.

It's a good piece. It's measured. It's not hype-coded. And it's a useful signal for the market: the trade press doesn't write about AI in tribal operations unless tribes are actually spending on it.

A few specific things they got exactly right.

What landed

Operational leverage, not transformation. The article frames AI as a way to reduce repetitive administrative work — not a replacement for the people doing it. Skokomish Indian Tribal Enterprises gets featured for using AI-powered accounting to automate invoice handling so staff can focus on analysis and operations. That's the right frame. We've been telling clients the same thing for two years: AI is a workflow hire, not a person replacement. The orgs that get this right end up with smaller teams doing higher-value work. The orgs that don't end up with shelfware and an annoyed CFO.

Tribes want to keep their information to themselves. Several people in the article talked about not wanting their information sent into big public AI systems. This is the right concern, and it doesn’t get enough attention. When we set up AI tools for tribes, the first question is always: whose system is this running on, who can see the information, and who keeps it when the work is done? Public AI services that learn from what you send them are not okay for sacred information, member records, or even regular tribal data. The article calling this out in print is helpful — it gives buyers the words to bring to their board.

Real examples for tourism work. For tourism-related work, the article lists understanding your visitors, marketing tailored to specific groups, tracking who keeps coming back, and planning ahead. That’s not a generic AI list — that’s the actual list of what a tribal tourism program needs to think about. We’ve already built most of these for AIT, and the other ones on the list are next.

People first, AI second. The article keeps coming back to this: AI works when it helps tribal staff and tribal storytelling, not when it tries to replace them. That difference is why some projects get used and others get rejected by the staff six months in. The rules about what can be shared, what stays inside the tribe — those aren’t edge cases. They’re the most important part of every decision.

So far, so good. The article validates a market that we've been working in for the better part of a decade. If you read it as a tribal exec thinking about AI — go read it. It's an honest read.

What they underplayed

Here's the part of the picture the article didn't quite reach:

Most of these uses are ongoing work, not one-time projects.

The way the article reads, AI sounds like a series of separate projects: set up the accounting tool here, the analytics tool there, the grant-writing tool somewhere else. Each one with a beginning and an end.

That works for the first AI thing an organization does. It doesn’t work for the second one, or the third, or year three when six AI tools are running across four departments and somebody needs to keep them working together.

This is what most consulting engagements get wrong. They sell a project. They build it. They leave. Then six or twelve months later, the AI changes, costs go up, the software underneath shifts, or federal rules change — and there’s nobody around who knows how to handle it. What looked great in the board meeting turns into unused software faster than the system the organization was already frustrated with.

What actually works — and what we’ve seen work over ten years with AIT — is treating AI as something you keep working on rather than as a series of one-time projects. Each individual piece gets built quickly (most of our weekly summaries take less than four weeks from start to finish). But there’s a senior tech person who stays around — keeping track of what’s running, what’s broken, what needs replacing, what should be added.

That’s the gap the article doesn’t quite name. Tribes thinking about AI for the first time usually ask "what should we build?" The more important question, after the first build, is: "who’s going to keep it useful?"

What to actually take from this

For tribal leaders thinking about AI right now, three things from the article are worth taking seriously:

1. Pick one specific problem to solve, not a general "AI strategy." Find the thing that’s eating your staff’s time and tackle that one. Accounting, grants, the weekly leadership summary, understanding visitors — pick one.
2. Make protecting your information a real requirement, not a footnote. Get it in writing with any company you work with. Where does your information live, who can see it, how does deletion work. Vague answers are a red flag.
3. Decide who’s going to keep it useful before you build the first thing. Either build the skills inside your own team, or set up an ongoing relationship with a senior tech person who stays involved. Organizations that build it and walk away end up with unused software. Organizations that build it and stay engaged end up with something that gets better over time.

The article reads the situation correctly. The wave is real. The organizations that move early will have an advantage. The question, for any tribe reading it, is whether to handle this through a series of separate projects — or to set up the kind of ongoing tech support that makes each project build on the last.

We’re partial to the second answer. The math works. Ten years with AIT is the proof.

If you're a tribal organization or indigenous-serving nonprofit reading this, a 15-minute conversation is the cleanest first step. We don't pitch on those calls — we look at where AI actually fits in your operation, and where it doesn't.