The Rules Were Always Temporary

There's a scene in Hackers—the 1995 film that probably introduced more people to the idea of digital culture than any actual computer—where a kid breaks into a system not because he knows exactly what he'll find, but because the act of looking is the point. The curiosity is the method. The system doesn't own you if you're willing to go somewhere it didn't expect.

I think about that energy a lot these days. Not the breaking in—but the disposition behind it. Exploratory, irreverent, genuinely thrilled by the unknown. It's the same energy that produced some of the most vital creative work of the last thirty years. And it's exactly what the design industry needs to rediscover right now.

Before the rules, there was chaos.

In the mid-nineties, while most of the industry was still figuring out what interactive even meant, a handful of studios and artists were already doing something else entirely.

Tomato—the collective that shaped so much of the era's most arresting visual communication, and the creative force behind Underworld's visual identity—weren't operating from a rulebook. They were treating design as a live experiment: typographic systems that felt like they were short-circuiting, imagery that refused to sit neatly between art and commerce, work that was intentionally unresolved. The Designers Republic, out of Sheffield, were building a visual language for electronic music culture that looked like corporate communication dismantled and reassembled wrong—deliberately, precisely wrong.

These weren't studios that had cracked a process. They were studios that understood something more durable: that the work happens at the edge of what you know, and you find it by going there before you're ready.

OK Computer didn't happen because someone engineered an album to define the anxiety of the late twentieth century. It happened because Radiohead deliberately abandoned conventional studios, treated uncertainty as material rather than obstacle, and let the process find the work rather than the other way around. Jonny Greenwood described it as trying to recreate brilliant records and missing. That missing—that productive failure—is the whole point. And it points to something I keep coming back to: the most generative moments in design history happened when the tools outran the rules. We are in one of those moments right now.

How design and development drifted apart.

To understand why this moment feels so charged, it helps to understand what happened in the decade that preceded it. As digital products grew more complex—microservices, component libraries, design systems, multi-platform deployment—the distance between design and engineering grew with them. Not because anyone wanted it to, but because the stack demanded specialisation. You couldn't hold it all in your head at once, so the disciplines formalized around what each role could reasonably own. 

Figma became the border. Designers worked up to the handoff. Engineers worked from it. The gap between a design file and a shipped product—technically, culturally, linguistically—widened into something that required its own infrastructure to manage. Whole job functions emerged just to bridge it. 

The aesthetic consequences were real, if harder to measure. When design and engineering operate in parallel rather than together, work tends toward the generic. Component libraries optimized for consistency produced interfaces that looked like every other interface. The pressure to design within what was technically feasible—as defined by someone else, interpreted at arm's length—quietly eroded the space where genuinely surprising work could happen. Design got more rigorous and, in many places, less interesting. 

The designers I came up with in the early two thousands would have found this strange. Back then, the roles hadn't separated as widely yet. Macromedia Flash was the closest thing we had to a universal creative instrument—a single tool where you could design, animate, write code, and edit sound in the same session, often in the same breath. It collapsed the distance between imagination and execution in a way that felt almost reckless. You might be building an interaction system in the morning and composing the audio that accompanied it by afternoon. The line between making something and building it was blurry in a way that felt generative—even necessary.

Nobody had agreed on where design ended and development began, which meant the space between them was open and alive.

AI is closing the gap.

What I'm watching happen with AI—and what leading this transformation at Huge has made very clear to me—is that the wall is coming down again. Not because the technical complexity has gone away, but because AI is absorbing enough of it to let designers get closer to the material again.

A designer who can prompt their way to a working prototype, who can generate and test interaction logic without waiting for an engineering sprint, who can move fluidly between the visual and the functional — that designer is operating in a fundamentally different relationship with the product. The handoff doesn't disappear, but the creative dependency on it does. And when design and development start speaking the same language again, work gets interesting again. 

This is the convergence that gets underreported in conversations about AI and design. The discourse tends to focus on what AI replaces or what it threatens. What it's actually doing, at the level of creative practice, is restoring a kind of fluency that the industry spent fifteen years engineering out of the process.