Every few years, a new tool triggers a version of the same conversation: will this replace designers? Desktop publishing, Photoshop, responsive frameworks, no-code platforms -- each one was going to eliminate the need for design expertise, and each one instead changed the nature of what design expertise meant. AI is doing the same thing, but at a different scale and a different speed. The question is not whether your job will exist. It is whether the particular version of it you currently do will still be valued when the dust settles.

What AI commoditises

Pixel execution is the clearest casualty. If your value proposition is that you can produce polished layouts quickly and consistently, AI tools do that faster and cheaper. This has been coming for years -- the Figma component model already automated much of the assembly work that used to differentiate fast designers from slow ones. AI completes that arc. The designer whose value lived in their ability to move quickly in tools is in a genuinely difficult position.

Template thinking is the other casualty. If your approach to design problems is to find the pattern that most closely matches what you have seen before and adapt it, AI can now do that search faster and more comprehensively than any human. Pattern-matching to established UI conventions is a useful starting point, but if it is also your finishing point, the value of having a human do it is declining rapidly.

Volume-based output -- generating many variations, producing many options quickly, iterating rapidly within a defined direction -- is also increasingly automatable. If a significant part of your working week is spent on output volume rather than on the judgements that determine what gets produced, that part is under pressure.

What AI amplifies

Judgement is the first thing AI amplifies, because it makes the gap between good and bad judgement more visible faster. When execution is cheap, decisions about what to execute are exposed immediately. A designer with sharp instincts about what to test, what to prioritise, and what the business actually needs will surface those instincts in every AI-assisted artefact they produce. A designer without those instincts will produce polished work that misses the point just as quickly.

Taste is amplified in the same way. The ability to look at ten AI-generated options and know which one is right -- not based on personal preference but based on what the product needs, who the user is, and what the context demands -- is a genuine skill that AI does not have. Taste is accumulated through years of looking at what works and what does not, in enough different contexts to build a calibrated model. That model is yours.

Commercial strategy is the third amplifier. Designers who understand the business model, the competitive landscape, and the commercial implications of their design decisions are valuable in ways that are very difficult to automate. The designer who can say 'this approach reduces support costs but creates onboarding friction that will show up in sixty-day retention' is operating at a level AI cannot reach, because it requires contextual business intelligence that AI does not have access to.

Stakeholder navigation: the non-automatable skill

Navigating stakeholders -- understanding their motivations, managing competing agendas, building the trust and political capital that allows good design to survive the organisational process -- is human work in a way that is not about to change. This is not a design skill in the traditional sense, but it is a skill that determines whether design has impact. AI makes it easier to produce work. It does not make it easier to get work approved, funded, or built.

The designers who should be worried are those who have been protected by execution speed and tool mastery without developing the judgement, taste, and commercial acuity that make design genuinely valuable. The transition AI is driving is forcing that value to the surface in a way it was easy to avoid before. There is nowhere left to hide behind 'I'm the one who knows how to do this in Figma'.

How to position yourself for this shift

The designers who should lean into this moment are those who already lead with thinking rather than making. If you have always been the person who reframes the brief, challenges the solution space, and brings commercial perspective into the design conversation, AI gives you more time to do that work and less time spent on execution. Your leverage increases.

For those who recognise the gap between where they are and where they need to be: the path is deliberate. Invest in the skills AI cannot replicate -- systems thinking, commercial acuity, research rigour, stakeholder communication. Use AI tools to remove the work that does not require those skills, so you have more space to develop the ones that do. The shift is happening regardless. The only variable is whether you are ahead of it or behind it.