TALAMANA · THE AI LITERACY MAP FOR ARCHITECTURE AND DESIGN · Judgment · AGE 17—19 · FACTUAL · DURABLE

It changes your second question

A plausible first output narrows the search before your own idea has formed.

The idea

Once a plausible answer is in front of you, it becomes the thing everything else is measured against. Your second question is shaped by it. You ask for "more like this, but…" instead of asking what else was possible — that is what the Lab sees in its studio logs. Psychologists call this anchoring, and it works on numbers, images and ideas alike. In design search it is expensive, because the first output arrives in seconds, before your own idea has had time to form. The search narrows before it has begun.

Why it matters

What you ask next decides what you find. If the machine's first draw writes your second prompt, the machine is steering the search and you are only holding the wheel.

See it in the studio

A student asks for "a community library on a sloping site" and gets a cantilevered box with a glass front. Every prompt after that says "more like this, with…". Forty images later the slope has never been explored. The site's one real idea was abandoned in the first ten seconds.

Watch for this

The word "more" at the start of your prompts. "More of this." "More like that." Each one is an anchor speaking.

Try it

Run a brief once. Before writing the second prompt, write three questions that have nothing to do with the first result: about the site, the programme, the section. Then prompt from one of those. Compare where the session went against where it would have gone.

Prove it

Show a session transcript and mark the exact prompt where the first answer set the direction for everything after.

How it works

Anchoring was described by Tversky and Kahneman in the 1970s. An initial value, even an arbitrary one, pulls later estimates towards itself. The design version has now been measured with generative tools. In a 2024 experiment with sixty participants in a visual ideation task, people who used an image generator while ideating fixated more on an initial example and produced fewer ideas, with less variety and lower originality, than people who worked without it. That study measured the fixation, not the wording of anyone's second prompt; the "more like this" pattern is the Lab's observation, and you can check it in your own logs. FIRST IMAGE, one ring up, takes the same finding to the studio day: when to let an image into the room at all. The counter is metacognition, watching your own thinking while you think, which is why this card sits beside the Lab's cards on prompting as thinking about thinking.

What this idea builds on

What this idea opens up

Sources

Open this idea on the map · The complete map · Logika · RBDS AI Lab, India · revised every edition.

Age grows from 11 at the centre to 22 at the edge, and six sectors show the learning strands. Tab into the map and the arrow keys step from idea to idea, following the links where there is one. Enter opens the idea under the cursor, and E reads out its links and the reason recorded on each. Press slash for Search, question mark for the full key list, and Escape to leave. Open Ideas for the complete readable list, including what each idea builds on and what it opens up.

LOGIKA · RBDS AI LAB INDIA
ON-RAMP · AGE 11 · FIRST ENCOUNTERS, NOT GATES — IDEAS · — DEPENDENCIES
DONE
OPENS NEXT
SOLID — STANDS ON