TALAMANA · THE AI LITERACY MAP FOR ARCHITECTURE AND DESIGN · Studio Practice · AGE 19—22 · FACTUAL · EVOLVING

From chatbots to studio systems

The line from chatbot to studio system changes how long the machine acts without you.

The idea

A chatbot answers when asked. A custom assistant answers with your studio's material loaded. An agent works through a sequence — searches, drafts, checks, files — inside a loop someone designed. A studio system is several of these with rules between them, running over time. Read the four as one line and each new product launch becomes a point on it, not noise. What changes along the line is not cleverness. It is how long the machine acts without you, and so who answers for what it did.

Why it matters

"Should I use AI?" is four different questions at four points on the line, and the fourth is a question about responsibility, not tools.

See it in the studio

Asking a chatbot to summarise a bye-law is a one-off. A standing system that reads every new drawing against the bye-laws and flags possible breaches is a colleague who never sleeps and never signs. Who checks its flags? Who answers for the one it missed?

Watch for this

Treating an agent's run like a chat reply. A reply is read once and judged. A run touches files, sends mail, fills tables, and the judging has to be designed in before it starts, not applied after.

Try it

Take one task you do weekly — a precedent search, minutes, a drawing check. Write it four ways: as a chat question; as a question to an assistant holding your files; as an agent's sequence; as a standing system. Mark where a human must sign at each.

Prove it

Explain what changes in who is responsible when AI acts over time instead of answering once — using one studio task as the example.

How it works

The mechanism is the loop: a model proposes an action, a tool runs it, the result comes back, the model proposes the next. Published practitioner guidance recommends starting with the simplest loop that works and adding autonomy only where it earns its place. EVOLVING because the line is still being drawn. What counts as an "agent", and how far it runs unsupervised, shifts with each release, and the responsibility question is being settled in practice before it is settled anywhere else. The Lab's own book chapter on agents — its own material — is the fuller treatment.

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