I’ve been trying to get my head around AI again. I do this regularly now, not because I’ve figured it out, but because it keeps shifting every minute. Every time I feel like I’ve got a handle on what it means for work, something shifts. A new LLM update drops. An engineer writes that they haven’t written code themselves in three months. I mentioned the rushing river in another piece of writing but it is a bit like trying to read a river by staring at one patch of water. I keep circling around what it means to be effective right now — less the tools themselves — when everything is changing faster than our professional identities can keep up with.
A while back I came across a talk by Jenny Wen that stuck with me. Her argument, roughly, was that the design process is dead — or at least that we need to let go of it in favour of actually working with the material: coding, prototyping, shipping fast, building closer relationships with engineers. I found myself nodding along, because there’s real truth in it. The slow, stage-gated design process was always a bit of a fiction anyway. We were never as linear as the diagrams suggested.
However, I’d add something to her framing. Before we pick up these powerful tools and start generating, we still need to be clear and intentional about what we’re actually trying to do. The risk on the other side of the “move fast and build with AI” mantra is what I’ve started thinking of as vibe-coded slop — decent enough looking outputs that haven’t been thought through, produced at speed and without much sense of purpose. The tools make it very easy to make things but that’s not the same as making the right things.
There is a version of AI-assisted design that’s genuinely exciting and a version that’s just noise at velocity. The difference, I think, is intention.
What Gen Alpha is telling us
Here’s what I find interesting as a counterpoint. Gen Alpha — the cohort growing up entirely inside this technological moment — is showing early signs of scepticism. Not rejection, but a more critical and selective relationship with AI than you might expect from digital natives. Nostalgia is one of the highest trending tags on TikTok. De-influencing is real. There’s a counter-movement with young people to disconnect and find things that feel human-made and earned, pushing back on the deluge of content that AI can produce faster than anyone can meaningfully consume.
I find this genuinely fascinating, and not just as an interesting nugget of cultural change. I think it’s saying something about what gets lost when everything is frictionless and fast. The things that feel meaningful — a well-observed piece of writing or a handmade object — often carry the trace of the effort behind them. Take away the effort and you remove something else: harder to name but easy to sense.
What stays durable
I’m not making an argument against AI in design. I’ve been using Cursor, Claude, Framer and playing with what’s possible, experimenting and sometimes being genuinely dizzied by what these tools can do. I don’t think there’s a moment to lose in staying with the pace of things. Falling behind feels like a real risk and not an abstract one.
But I do think the designers who will navigate this well aren’t just the ones who adopt AI fastest. They’re the ones who stay strategically clear, who hold onto the skills of human connection and synthesis that these tools can’t replicate, and who bring a real point of view to what they’re building before they start building it.
What’s becoming the baseline — AI fluency, faster iteration, closer collaboration with engineers — was a differentiator a year ago. I think that window is closing now. What stays durable, I suspect, is the judgment underneath. Knowing what matters. Understanding the people you’re designing for. Being intentional about what you put these powerful tools to work on.
The river is still fast. I’m still wading in. But I’m holding onto a few things as I go.