LO
Architecture
Trains you to turn a brief, a site and a budget into something that can actually be built — and to be the person who is answerable when the drawing meets the ground.
Generate a set of scheme images for one of your projects with a model, put them beside your own, and write down what a critic could say about yours that they could not say about the generated set. That sentence is what your visualisation hours are still buying you, and if you cannot write it, that is the finding.
A single number for a whole major would hide the thing that matters: this degree trains several separate competencies, and they are not all moving in the same direction. Automation acts on tasks, so any assessment lives on the occupation pages below — not here.
Written for undergraduate and master's architecture programmes in mainland China and Singapore, on the path toward registration. Interior design, landscape and pure visualisation studios diverge — visualisation in particular is far more exposed, and the occupation page says so.
What this degree actually trains#
Not the course list — the competencies underneath it, and whether each one is worth more or less than it was.
Resolving constraints that contradict each other
Worth more than beforeHolding budget, code, site, structure, programme and a client's wishes at once, and finding the move that gives up the least — then defending why that was the thing to give up.
Generating plausible options is now cheap, which moves the value to choosing between them under constraints that cannot all be satisfied. That choice carries liability, and liability is not something a model can hold — which is also why the approvals and site tasks on the occupation page are assessed as human-led with evidence behind them.
Drawing something that can actually be built
HoldingConstruction documents and details: the drawings a contractor prices, orders from and builds to, where an ambiguity becomes a change order.
This is the occupation's most evidenced task and it is assessed as being augmented rather than automated: tools now draft and check the routine parts, while the decisions about what the detail has to survive still sit with a person. The drawing is not going away; the hours spent producing it are.
Producing the images that sell the idea
Worth less on its ownRenderings, presentation boards and the visual case that gets a scheme approved by a client or a committee.
This is the task the occupation page marks as automating, and it is the one students most often practise hardest because it is what a portfolio is judged on. Image generation reached presentation quality first; the skill that survives is knowing which image answers the objection in the room, not producing it.
Reading the code and carrying the compliance
Worth more than beforeKnowing what the building code, fire rules and planning conditions actually require, and being the signature on the submission.
Automated plan checking is arriving — this site records one city reporting permit decisions averaging 32.5 days with its AI plan-check software against 73 without. A machine that checks compliance faster does not remove the person who is accountable for it; it moves that person's time from finding errors to answering for the ones that remain.
Explaining something that does not exist yet to someone who is not an architect
HoldingGetting a client, a resident or a committee to understand what they will actually get, including the parts they will not like, before anything is built.
Neither displaced nor newly scarce. It was always the part of the job that decided whether a scheme survived, and nothing about generated imagery changes who has to stand in the room when the objection comes.
Where it can lead#
4 directions, never one. Each is written out in full on its own page — what your training reuses, what graduates typically lack, the entry reality, and one thing to test this term.
What to add outside the classroom#
This is about what graduates commonly lack in practice — not a claim that your school failed to teach it.
Time on a construction site, in any capacity. The studio teaches design and assesses drawings; almost nothing in a degree teaches what happens to a drawing once it is built from, and that is where the judgement this profession pays for comes from.
One real cost exercise. Knowing what a square metre, a window or a stair actually costs separates a scheme that can be built from one that only renders well, and it is rarely graded.
Deliberate practice at the compliance side — reading an actual code clause and working out what it forbids. It is the least glamorous part of the degree and the one that is becoming more valuable rather than less.
This term#
One or two actions, each producing something you can show someone. Not a reading list.
Generate a set of scheme images for one of your projects with a model, put them beside your own, and write down what a critic could say about yours that they could not say about the generated set. That sentence is what your visualisation hours are still buying you, and if you cannot write it, that is the finding.
Ask one architect more than five years into practice how their week divides between drawing, coordinating and answering for things. The ratio is rarely what students expect, and it is the ratio the assessment on this site is about.
Common questions#
Look at which task is moving before deciding, because they are not moving together. On the architect page, visualisation is assessed as automating, construction documentation as being augmented with evidence behind it, and approvals, coordination and site decisions as human-led — also with evidence. If what drew you to architecture was making the images, that is the exposed part. If it was resolving a site and being answerable for what gets built, the degree still leads there and the route is long for reasons that have nothing to do with AI.
It already generates schemes and images; what it does not do is carry the consequence. The evidence on this site points at the compliance layer moving first — one city reports permit decisions averaging 32.5 days with AI plan-check against 73 without, and corrections per application falling from 23.5 to 7.7. Note what that record does and does not establish: it is one city, residential only, about forty self-selected applications, and the department's own figures. Faster checking does not remove the signature on the submission.
That is the right question for this major specifically, because the sunk cost is larger here than almost anywhere else and finding out in year four is expensive. The honest framing is that the length is a regulatory fact rather than a market one — the years exist because someone must be accountable for buildings, and that accountability is the same reason the human-led tasks on the occupation page are assessed as they are. The risk worth weighing is not automation; it is whether you can sustain a long, low-paid early stage in which the hours skew toward exactly the tasks that are changing.
It is still what is asked for, and it is also the task this site assesses as automating — which makes a portfolio of images a weaker signal every year, because the reviewer knows what they cost to produce now. What does not weaken is evidence of judgement: a scheme with the constraint that shaped it written down, the option you rejected and why, a detail you priced. That is harder to assemble and considerably harder to generate.
Site time, one costed element, and one scheme you can explain as a sequence of decisions rather than a set of images. Between them they cover the three things the degree does not assess and the profession pays for: what happens after the drawing, what things cost, and why you chose this over the alternative. All three are also the parts of your portfolio a reviewer cannot assume were generated.
Method#
Assessments live on tasks, not on majors. Follow any direction above to its occupation page to see which tasks are changing, how strong the evidence is, and what it does not yet show.