The Return of the Human Hand — and What AI Can't Fake

Here's something I didn't expect to be writing about in 2026: watercolours.

Not as a retro novelty. Not as a "vibe". As a deliberate, strategic creative choice — made by designers and artists who know exactly what digital tools can do, and are choosing something else anyway.

Today's stories paint a picture (literally, in some cases) of a design world that's quietly reasserting the value of the human hand. And if you're a designer figuring out where AI fits in your practice, this matters more than you might think.

Craft is having a moment — and it's not nostalgia

Two stories caught my eye today. First: the team behind Croak, an upcoming game from Woodrunner Games, is building their entire visual identity around traditional hand-drawn animation. We're talking frame-by-frame drawing, squash and stretch, carefully timed keyframes — the stuff animators have been doing since the 1930s. Their argument? It delivers a warmth and personality that purely digital methods can't match.

Second: digital artist Roberto Gatto has turned to physical watercolours — inspired by Miyazaki and Studio Ghibli — to build the world of his RPG. Not watercolour brushes in Procreate. Actual paint on actual paper, scanned into an inherently digital format.

Neither of these creators is a Luddite. They're making conscious choices about where human craft adds something irreplaceable.

And that's the key word: irreplaceable.

We're surrounded by AI image generators that can produce polished artwork in seconds. But what these projects demonstrate is that "polished" and "alive" are not the same thing. There's a texture to hand-drawn work — imperfections, timing, the slight wobble of a real brush — that audiences respond to on a gut level. AI can approximate the look. It can't yet replicate the feel.

For designers wondering whether craft skills still matter in the age of generative AI: yes. Emphatically yes. Not because AI is bad, but because the gap between AI-generated and human-made is becoming a meaningful design choice in itself.

Making good design inevitable

Meanwhile, on the more practical side of things, there's a smart piece doing the rounds about using Figma variables to test font scaling for accessibility — right inside your normal design workflow.

The idea is straightforward: instead of treating accessibility as a separate checklist you tackle at the end (or, let's be honest, forget about entirely), you embed scaling tests directly into your Figma setup using variables. Change a variable, see how your type holds up at different sizes. No context-switching, no separate tools, no excuses.

What I love about this approach is the philosophy behind it: make accessible design "almost inevitable rather than optional."

That's a principle we should be applying to AI tools too. The best AI integrations aren't the ones that ask you to do something extra — they're the ones that make better outcomes the path of least resistance. If your AI-assisted workflow makes it easier to skip accessibility testing, something's gone wrong. If it makes it harder to miss, you're onto something.

This is where I think the conversation about AI in design needs to go. Less "what can AI generate?" and more "what does AI make inevitable?" Are your tools nudging you toward better work, or just faster work?

Designing for feeling, not just function

One more story worth your time: an article comparing how anime and Western superhero films handle emotional pacing — and what product designers can learn from both.

The argument is that anime tends to let emotions flow and coexist (joy alongside melancholy, tension alongside wonder), while superhero stories tend to compartmentalise emotions into distinct beats. Both work. But they create fundamentally different user experiences.

Translated into product design, this becomes a question about emotional architecture. How does your product guide someone through a feeling? Is the transition from onboarding to first success jarring or smooth? Does your error state kill momentum or hold space for recovery?

This is the kind of thinking that AI tools are nowhere near replicating. An AI can generate a modal, a page layout, a colour palette. It cannot yet reason about the emotional rhythm of a five-screen flow. It doesn't understand that sometimes you need a pause between two interactions — not for usability, but for breath.

If you want to future-proof your design skills, this is where I'd invest. Not in prompt engineering. In emotional intelligence applied to interfaces.

Tool spotlight: Wacom Bridge goes mobile

Quick one for the hardware-inclined: Wacom is pushing forward with its Bridge platform, aiming to let creatives use their pen display ecosystem from virtually anywhere. If this delivers on its promise, it could genuinely change how illustrators and designers work on location. Worth keeping an eye on — especially if your workflow already involves a Wacom setup and you've been frustrated by being tethered to a desk.

So what's the takeaway?

Today's stories aren't anti-AI. I want to be clear about that. But they collectively point to something important: the most interesting design work right now is being made by people who understand what AI can do and know when to reach for something else.

Craft isn't dead. Emotional nuance isn't automatable. And the smartest tool integrations are the ones that make good practice the default, not the exception.

If you're building your skills for the next few years, here's my advice: learn to use AI well, absolutely. But don't let it atrophy the things that make your work distinctly yours. The wobble of the brush. The rhythm of the flow. The judgement call that no model can make for you.

Those are your superpowers. Protect them.