TALAMANA · THE AI LITERACY MAP FOR ARCHITECTURE AND DESIGN · Studio Practice · AGE 20—22 · FACTUAL · DURABLE
Budget cost and time
The cost of a batch is mostly the cost of reading it.
When to use
Planning any project stage with an AI step in it, and pricing any fee that assumes one.
The method
Before running, estimate two numbers. Money: calls × rate, at the quality you will actually use, plus the re-rolls you already know you will make. Time: not generation time but turnaround — generation plus curation plus verification plus the rebuild when something drifts. Write both into the schedule as named line items. Set a gate: above a stated spend or a stated day, stop and decide rather than drift. After the stage, record the actual beside the estimate. That ratio is your calibration for next time.
Watch for this
Budgeting the generation and forgetting the looking. Fifty images take an hour to make and a day to judge properly. The cost of a batch is mostly the cost of reading it.
Try it
Estimate the money and the hours one design task's AI stage will cost before you run it. Run it. Compare. Then, on paper, reshuffle a tight timeline when the engine you chose turns out too slow or too costly for the deadline.
Prove it
Show one project schedule where the AI stage appears as named line items with an estimate, a gate, and the actual recorded beside it.
How it works
Per-call prices and generation speeds change monthly, so the method is a habit, not a price list. There is a second cost the bill never shows: the energy and compute behind each call. That is why the hidden-layers essay belongs here. The running cost of a studio system varies too widely to give honestly as one monthly figure: archive size, model choice, frequency of use and hosting all move it.
What this idea builds on
What this idea opens up
- Nothing yet names this as a foundation.
Sources
Open this idea on the map · The complete map · Logika · RBDS AI Lab, India · revised every edition.