We are starting to field questions from clients on how we can/will/should use AI when building out a website. Both in terms of process and deliverables and in terms of embedding AI features into clients’ websites. We are in somewhat early days regarding AI, and these are big questions.

And like many people, we have mixed feelings about artificial intelligence.

We have concerns about how we, as humans, use technology now and about how we will use it in the future. We worry about humans outsourcing too much thinking, writing, and decision-making to machines. We’re uneasy about the pace at which these tools are being deployed, often without clear guardrails, transparency, or long-term consideration of their social, physical, and environmental impact. We wonder whether humanity is better off, on the whole, for having these tools or whether they represent the beginning of the end.

Those concerns are real, and we don’t dismiss them. We share them. They even keep us up at night.

At the same time, AI tools already exist, they are widely available, they’re improving quickly, and they’re already reshaping how work gets done across nearly every industry, including ours – in some ways that are already clear and in some ways that are still emerging.

So we are thinking about how these tools impact our work and taking a deliberate, values-driven approach to AI: we use AI where it meaningfully improves our workflow, benefits our clients, and supports better outcomes, while vowing to always stay firmly grounded in human judgment, accountability, and care.

We’re not the decision-makers about the future of AI. But we are responsible to our clients.

We didn’t choose for AI to become embedded in modern work. Governments, corporations, and technology platforms made those decisions at a scale far beyond any individual studio or client.  What we can decide is what our values are and how we respond to a changing landscape.

As a small, independent digital agency, our responsibility is to the organizations that trust us with their websites, budgets, and missions. If tools exist that can:

• reduce unnecessary labor,
• make our process more efficient,
• help us surface blind spots,
• or allow us to deliver better results at a more sustainable cost,

Then we feel an obligation to evaluate and, where appropriate, use those tools on our clients’ behalf.

Not recklessly. Not automatically. And not without human thinking firmly at the helm.

What a thoughtful approach to using AI means for us

When we say we use AI, we don’t mean that we hand a website over to a robot and walk away.

We use AI much the way we use other professional tools: as assistive technology, not as a replacement for expertise or responsibility.

In practice, and as examples, that means AI may help us:

• surface the cause of a bug
• gather resources
• synthesize notes from meetings or discovery sessions
• identify patterns or gaps in content and site structure
• suggest clearer ways to organize information
• generate alternative wording or headlines for discussion
• flag potential accessibility or usability issues
• assist with quality assurance before launch

In all cases:

• humans make the decisions
• humans review and edit the work
• clients approve the final result

AI won’t “decide” what your website says, how it looks, or what it prioritizes. We (MIGHTYminnow, in collaboration with our clients) do the thinking, using experience, context, and judgment.

What we don’t use AI for

Just as important are the lines we don’t cross.

We don’t use AI to:

• mass-produce generic content
• replace client voice, expertise, or lived experience
• bypass thoughtful strategy or editorial review
• promise rankings, traffic, or “AI optimization hacks”
• embed AI features on sites unless there’s a clear, user-centered reason
• Implement “accessibility improvements” that don’t actually improve the experience of humans

We’re not interested in novelty for novelty’s sake. If a tool adds cost, complexity, or risk without a clear benefit, we don’t use it.

Why this matters for affordability and quality

By reducing time spent on repetitive or mechanical tasks, we can:

• focus more energy on strategy and problem-solving
• spend more time collaborating with clients
• keep projects within realistic budgets
• avoid passing inefficiencies on as higher costs

In other words, thoughtful use of AI can help us do less busywork and more meaningful work – which benefits everyone involved.

Holding complexity, not pretending certainty

We don’t think the right response to AI is blind enthusiasm, nor do we think total rejection is reasonable given where we find ourselves.

We believe it’s possible, and necessary, to hold complexity:

• to acknowledge the risks
• to advocate for sensible guardrails and regulation
• to stay curious but cautious
• and to make grounded, ethical choices within the reality we’re operating in

We’ll continue to revisit how we use these tools as they evolve. If something stops serving our clients well, we’ll stop using it. If better standards emerge, we’ll adopt them.

For now, our approach is simple:

Where AI helps us think more clearly, work more efficiently, and deliver better results for our clients, we will use it – thoughtfully, transparently, and with humans firmly in charge.

If you ever have questions about how we’re using AI on your project, we’re always happy to talk it through.

One last note: we paid for ChatGPT briefly, but we cannot give any more money to OpenAI as their values are wildly divergent with our own.