Five Principles for Adopting AI Responsibly in Tourism

Post 3 of 3 in a series on responsible AI adoption for destination marketing organizations (DMOs) and tourism leaders.

After two posts, here's where you stand. You're not behind on AI. You're not naively optimistic about it either. You can see its costs across environmental, social, and governance dimensions, and you understand that these need to be weighed honestly against AI's benefits to determine if adoption is right for you. You're ready to adopt AI in your organization, deliberately and responsibly. The question is how.

That's what this article is for. What follows is five principles for adopting AI responsibly. Each one is short enough to remember, durable enough to hold up as the technology keeps evolving, and concrete enough to act on this quarter.

1. Use AI deliberately

Two questions sit at the center of this principle. Should AI be used here at all? And if it should, what level of capability does the task actually need?

The reflexive move is often to reach for AI on every task, and to reach for the most powerful version of it. The deliberate move is to ask both questions before reaching. Not every problem needs AI. Of the problems that do, only a few need the most powerful model on the menu.

This principle does more than address the environmental costs (water, energy, emissions) outlined in post 2. Unreflective AI use also deskills work that humans should keep doing and it can produce content that flattens what makes your destination distinctive. 

Used deliberately, AI earns its place as a powerful tool.

In practice: ask whether AI is the right tool for each task before deploying it. When it is, match capability to actual need. Much DMO work is well-served by basic and mid-tier models, not frontier ones. Choose AI providers that publish their environmental footprints and goals.

2. Keep humans on consequential decisions

AI can analyze, suggest, draft, and execute. When the decision is consequential, a person decides. This holds whether AI is advising you or acting on your behalf. As agents take on more, the question of where humans stay in the loop becomes more important, not less.

What counts as consequential is for each organization to define. Customer-facing commitments. Content with reputational stakes. Decisions that affect staff or community. Anything irreversible. The principle isn't that humans review every output. It's that humans are the final word on what matters, and you've been explicit about what matters.

In practice: define your consequential decisions in writing. Build approval checkpoints into workflows. When AI acts agentically, hard-code action boundaries such as caps on what it can commit to, requirements for explicit consent on certain actions, and scope limits it cannot exceed.

3. Keep responsibility with people

Responsibility doesn't transfer to the technology. When AI produces output, a person remains responsible for it. The 2024 Canadian tribunal ruling against Air Canada made this legally explicit. It was always true ethically.

Every AI use in your organization has a named human owner. Not a team. Not a role. A person. That person is accountable for what the AI produces, including when the AI surprises them. The "AI did it" defense isn't available, and pretending otherwise creates legal exposure and erodes trust. This applies to vendor choice as well. When you bring in an AI tool, your organization is accountable for what that tool does, regardless of what the vendor's terms of service say.

In practice: for each AI tool or workflow, write down the owner. If you can't, that's the first thing to fix. The clarity you build internally is what protects you when external accountability arrives, whether that's an unhappy customer, a regulator, or a court.

4. Stay honest about AI's role

People deserve to know when AI is shaping what they encounter. Visitors talking to a chatbot. Staff whose workflows are being automated. Community members whose representation is being produced by AI. The level of disclosure varies, but the principle of honest disclosure stays constant.

Regulation across major jurisdictions is moving toward explicit disclosure requirements for AI-generated content, deepfakes, and chatbots. But beyond regulation, disclosure is how trust gets maintained. Audiences who learn after the fact that AI was involved tend to feel deceived, even when the AI did good work. Audiences who knew up front can engage on honest terms.

In practice: map your AI uses to the audiences each one touches. Define what each audience needs to know. Default to disclosure unless there's a specific reason not to.

5. Make AI decisions like a place steward

DMOs sit in a position other organizations don't. You're not just an internal adopter of AI. You shape how a destination shows up in the world, and you're accountable to staff, host communities, and the people the destination represents. AI decisions affect all three.

When AI changes what frontline tourism work looks like, you have both internal responsibility (transition planning, upskilling) and industry standing (advocating for sector-wide protections) to consider. When AI infrastructure expansion happens in your region, with data centers competing with residents and visitors for water and power, you have influence to advocate on whether the expansion fits the carrying capacity of your region. When AI generates content about the communities your destination represents, you have editorial responsibility (community review, OCAP/CARE/FPIC where applicable).

In practice: identify which roles in your organization are most AI-exposed and plan their transition. Take a position on infrastructure decisions in your region. Build community review into your AI content workflow. The role you already play as steward extends to AI decisions.

Why principles, not playbooks

A short while ago, AI was just an advisor. You asked it a question, it gave you an answer, you decided what to do with it. Now AI is also a doer. Agents like Mavis at Malaysia Airlines or the Sabre/PayPal/MindTrip booking system don't just answer questions; they take actions on customers' behalf. Booking. Refunding. Rescheduling.

That shift took less than two years. It's part of why a playbook approach to AI adoption fails. A playbook written for advisor AI doesn't cover agentic AI. The next shift will arrive faster than the one that just happened. By the time anyone writes the agent playbook, agents will already do something that breaks it.

Principles hold up under the rate of change we're experiencing because they describe how to decide, not what to do. The five above don't change when AI starts to act rather than advise. That said, these aren't the only principles you'll need moving forward. Borrow what works for your organization, break them, strengthen them, and revise regularly as the AI landscape changes.

How I work with this myself

A note on practice.

The metaphor I use to help me bring these principles into my own work more intuitively is that adopting AI isn't about learning new software. It's closer to onboarding a new team member. That framing changes what good practice looks like.

Use AI deliberately becomes a question of what tasks are appropriate for the new hire, especially in their first weeks. They don't get everything at once. They don't get the most consequential work first. I match the task to where they are, not to my hopes for what they could eventually do.

Keep humans on consequential decisions becomes a question of sign-off. I decide what the new hire can decide alone, and what has to go through me. Certain communications don't go out without my review, even when I'd bet on the content being correct.

Keep responsibility with people becomes a question of who the named manager is. That's me for every AI tool I use. If it produces something that goes out wrong, that's on me, not the tool.

The metaphor stretches at the edges. AI isn't a person. But for the daily practice of adoption, the new-hire framing keeps me honest in ways that treating it as software never did.

Where the work goes from here

Across three posts, this series has tried to do three things. Show that being behind on AI is an opportunity to adopt it responsibly. Make the costs of unexamined adoption visible across environmental, social, and governance dimensions. And offer a set of principles for adopting AI responsibly that hold up as the technology keeps evolving.

None of this resolves the underlying tension between using AI and its costs. That tension stays. The principles above don't make it go away. They offer one way to navigate it.

The discourse about AI is only going to get louder and more complicated as it becomes increasingly woven into our society. That discourse needs more voices from people who use AI responsibly and think carefully about how it should be shaped. Responsible adoption starts in your own organization, but it doesn't end there. The systemic change (regulation, infrastructure, accountability at scale) needed to complement individual practice requires advocacy that goes beyond any one person or organization. This technology is still young. There is real potential to shape how it gets integrated into society and how its impacts land. Join the conversation and lead with your actions.

If the series has done its job, you finish it with a clearer view of where you stand and the confidence to take the next step even as the ground keeps moving.

- TYLER ROBINSON

Tyler Robinson is the Founder and Principal Consultant of Tydal Consulting, where he helps organizations adopt AI responsibly and effectively. He brings over a decade of experience as a climate and sustainability strategist, advising places and organizations on how to operate more sustainably. He built the AI Environmental Impact Calculator to help organizations measure the impact of their AI usage.