Open a generative-AI tool, type “design a low-cost water filter for refugee camps” and, in seconds, you have concept drawings, a bill of materials and a rollout plan. Yesterday that required a term-long group project. Today, it’s the warm-up. The superpower now sitting on every student’s laptop is the ability to race from spark to prototype.
Trust me, I tried this and it works. The last time I coded anything, it was in BASIC in 1992. Last week, while I sat at a cafe, I had ChatGPT propose, design and implement a WordPress plug-in for my website, which allows students to map the syllabus against the time they have available to prepare towards their exam. From morning-shower-idea to implemented solution, in the time it took me to drink a flat white.
This changes every part of higher education, yet most curricula merely bolt a “How to prompt ChatGPT” slide on to last year’s lecture. Students, meanwhile, use the tech to vault over the skills we test them on. They no longer need weeks of coding workshops to build an app or statistics tutorials to run regressions – the machine does the legwork. Our value can no longer be the skills treadmill.
From skills to sparks
For centuries, universities have delivered scarce expertise. We stacked programmes like layer cakes: first theory, then practice, finally – if there was time – a sprinkle of creativity. Generative AI flips that order. Because routine skills are on tap, the bottleneck shifts upstream to ideation: spotting problems worth solving and framing them so the machine can help.
That demands divergent thinking, curiosity and ethical judgement – qualities our assessment regimes often squeeze out. We need to treat creativity as a core literacy, not a decorative extra. Don’t get me wrong, skills are not irrelevant – they just look different. Prompt craft, data stewardship and model critique replace manual citation and calculator drills. But they are means, not ends.
Studios, not lecture theatres
Ask three questions when redesigning a module: Where can AI shorten the distance from idea to outcome? How will students showcase originality, not output volume? When will they pause for ethical reflection?
When imagination is scarce, learning spaces must feel like design studios. Instead of teaching business law, droning on about director’s duties, I will get teams of students to build regulatory sandboxes for autonomous-vehicle start-ups. They will use legal-language models to draft compliance frameworks, then spend seminars debating what the bot missed and why it matters. Skills happen in the background – creativity sits centre stage.
By doing this, we can finally breach the disciplinary boundaries we long pretended to dislike. Creativity flourishes at the edges of disciplines. Pair literature students with computer scientists to prototype narrative-driven virtual reality; mix chemists with economists to imagine circular-economy start-ups. Set a shared brief, equip them with AI tools and a two-week sprint, then step aside.
The revelation is not how quickly they build things, but how naturally they trade perspectives once the translation burden sits with the machine. That habit of fluent collaboration is exactly what employers in an AI-saturated economy will prize – and universities, not tech platforms, are ideally placed to choreograph it. We preach interdisciplinarity, now is the time to do it.
Faculty as creative directors
Worried you may be left without a job? The bot will not replace the lecturer – it will replace the lecturer who only delivers content. Our new role is creative director: curating resources, modelling intellectual risk and orchestrating messy discussion.
Start by using the tools yourself. When students see you experiment – whether with an AI avatar explaining case law or a model drafting a syllabus – they learn that scholarship is exploratory, not finished. In trying to solve the challenge of student disengagement, I became an AI avatar designer. Creativity offered a pathway to solve my problem.
Traditional exams reward retention – coursework rewards hours invested. Both collapse when GenAI can write a passable essay in a minute. Instead, require an AI-assisted artefact plus a design log narrating the creative journey, missteps included. Mark the thinking, not the typing. To be honest with you, ChatGPT wrote most of this article, and it didn’t even take minutes.
A creativity contract
Here is my graduate attribute list for the AI era. My student will be a problem finder, who hunts for unmet needs; a prompt poet, who translates fuzzy ideas into machine-readable briefs; a critical friend, who interrogates model blind spots; a bridge builder, who links insights across disciplines. And finally, a moral navigator, who steers innovation toward the common good. Every attribute begins with a creative stance. That is the contract we must strike with applicants and employers: we will graduate people who can imagine what to do with limitless capability.
Potential students are already wondering what is the point of a master’s degree, when a bootcamp plus AI can let them launch a venture in six months. If universities cling to a skills-first identity, they will watch this exodus accelerate. Creativity, by contrast, ages well. The capacity to conceive, critique and reinvent will remain valuable long after today’s models are museum pieces.
Launch pads, not lecture notes
Let’s retire the phrase “future-proof skills”. There is no such thing. What we can future-proof is the human imagination. GenAI has handed us a supersonic airplane – our job is to help students pick destinations worthy of the ride.
So, ask your cohort next Monday: “What problem have you always wanted to solve?” Then build a classroom where turning that answer into a prototype is day-one homework. The creativity-first university starts there – before that aeroplane leaves without us.
Ioannis Glinavos is a senior lecturer in law at the University of Westminster.
If you would like advice and insight from academics and university staff delivered direct to your inbox each week, sign up for the Campus newsletter.