TALAMANA · THE AI LITERACY MAP FOR ARCHITECTURE AND DESIGN · Studio Practice · AGE 20—22 · POSITIONAL · HELD
An honest, strong portfolio
A hidden process reads as deception once discovered, and a shown one reads as authorship.
The dilemma
Your best spread has a generated hero image. The jury — or the employer — will ask how it was made. Hide the process and let the image speak, or show the brief, the iterations, and what you decided?
The choices
Present the final image alone and hope nobody asks. Or build the spread to show the judgment: the brief you wrote, the option set, the rejected ones and why, the one decision the machine could not make — and carry a one-line authorship note that says what was generated, what was drawn, and who directed it.
The consequence
A hidden process reads as deception the moment it is discovered, and it does get discovered — by a question, a reverse image search, a senior's eye. A shown process reads as authorship: the image becomes evidence of a chain of decisions, and the decisions are yours. Honesty is not a cost on the work. It is what makes the work yours.
The case
Two students submit renders of the same quality. One spread says "concept view." The other says: "generated from my massing model and material brief; thirty-one variants, three retained; final chosen for the west-light condition shown in the section opposite." The employer hires the second — not for the image, for the evidence of judgment.
Take it to crit
On a portfolio page, ask what the student decided that AI could not. Does the page show that, or hide it?
How it works
The authorship note is a small discipline with a large effect: it forces you to know, for each image, what you decided. The competition clause elsewhere on this ring is the same discipline imposed from outside. The portfolio is where you impose it on yourself first. Disclosure settles what you did. It does not settle who owns the result: in India the Copyright Act, 1957 counts drawings, plans and works of architecture as artistic works, has a line on computer-generated works written long before these tools, and says nothing yet about what a provider's terms grant you in an image the model made — and the Delhi High Court's July 2026 ruling on training was interim and concerned text. The author card on the Ethics and Provenance strand carries that question; this card carries the note.
What this idea builds on
What this idea opens up
- Nothing yet names this as a foundation.
Sources
- authorship follows judgment, shown not asserted
- curriculum-spine
- The Seduction of the Machine
- RIBA AI Report 2025
Open this idea on the map · The complete map · Logika · RBDS AI Lab, India · revised every edition.