What Does AI Stand For Now? Aesthetic Intelligence and Design's Reorganisation

The course is called AI → Aesthetic Intelligence. That arrow in the title is doing a lot of work. It forces anyone who reads it to pause for a fraction of a second and recalibrate. That pause is the point. But what exactly is being proposed inside it?

The question matters more than it might first appear, because the timing is specific. In 2023, Adobe announced that Firefly — its generative image suite — would be embedded directly into Photoshop and Illustrator. Not offered as a separate subscription, not tucked into a beta tab, but woven into the two applications that have defined professional design practice for roughly thirty years. That integration didn't just add a feature. It changed what it means to open those tools. A practitioner sitting down to Photoshop in 2026 is sitting down to an environment where generative image production is a native gesture, like a brush stroke or a layer mask. The workflow assumption that design is primarily the skilled execution of visual ideas — that the hand and the eye are what you're training — has been quietly reorganised underneath the profession.

Which is, interestingly, not the first time this has happened. The Bauhaus, founded in Weimar in 1919, was built on exactly this kind of structural reorganisation. Industrialisation had already changed the designer's actual job: it was no longer to make the object but to direct how the object was made at scale. The curriculum Walter Gropius designed wasn't a response to a new tool; it was a response to a new relationship between the designer and the means of production. The Bauhaus didn't ask "how do we teach people to use machines?" It asked "what is a designer actually for, now that machines exist?" That is a different question, and it produced a different kind of school.

The AI → Aesthetic Intelligence naming seems to be reaching for something similar. Not "here is how to use generative tools" but "here is what the discipline is actually for, now that generative tools exist." Aesthetic intelligence, as a phrase, implies that the scarce and trainable thing is judgment — the capacity to distinguish, to select, to know when something is working and why, to hold a position about quality in the face of infinite generative output. If a model can produce ten thousand variations of a visual idea in the time it once took a designer to produce one, the bottleneck shifts. It moves from production to discernment.

James Hurst's book Use Design To Design Change makes a related argument, though it arrives from a different direction. The book's central claim is that lasting value comes from purpose and values-rooted unconventionality rather than technical execution — that the thing which can't be copied is not the craft but the conviction underneath it. That's a claim about brands and organisations, but it sits in direct tension with the assumption, still embedded in most design education, that design is primarily a craft skill. If the craft is increasingly executable by generative systems, and if conviction is what actually creates durable value, then the course title starts to look less like a rhetorical trick and more like a genuine job description rewrite.

The honest question is whether aesthetic intelligence is teachable in the way that Photoshop is teachable. Tool fluency has a relatively clear pedagogy: here is the interface, here are the operations, here is the feedback loop that builds muscle memory. Judgment is harder to scaffold. You can expose people to a great deal of work, build a critical vocabulary, create conditions for taste to develop — but the Bauhaus took years to figure out how to do even that, and it had the advantage of a coherent philosophical framework and a faculty that included Paul Klee and Wassily Kandinsky. A short course, however well designed, is working against a different set of constraints.

And yet the alternative — continuing to teach design as though the primary output is skilled execution of visual assets — seems like it's answering a question that has already changed. The students sitting down to learn design in 2026 will spend their careers working alongside generative systems that are, by most measurable standards, already technically proficient. What they will need is not to out-execute those systems but to know what to ask them for, how to evaluate what comes back, and when to say that none of it is right yet. That is a different curriculum. It might be a more interesting one.

There's something in the Weird the Normal / Normal the Weird framework that feels adjacent here, even if it wasn't designed for this problem. James describes the actual skill not as landing on one side — pure strangeness or pure structure — but as the switching between them, maintaining a kind of productive tension. Aesthetic intelligence might work the same way: not a fixed sensibility that you acquire and then apply, but a capacity to move between the generative and the critical, between making and judging, between openness to what the tool produces and conviction about what the work is actually for.

What's genuinely unresolved is whether renaming the acronym is the beginning of that argument or the whole of it. The name is sharp. The question it opens — what is a designer actually for, now? — is the right question. Whether a course can teach the answer, or whether it can only point at the gap and hope people are curious enough to live in it for a while, is something worth sitting with.

Source: Rogue School