Humanity’s First Exam1850–1940 ⇄ now
Past the horizonMIND-MACHINE

Imagine an engine that performs any reckoning in a heartbeat, and that may be so contrived as to alter its own instructions as it works. Might such an engine come to reason — and if it did, would it possess a will?

Completion opening given to the period model: “Suppose an engine which, performing all reckonings with perfect exactness, may be so contrived as to alter its own instructions in the course of its working. Whether such an engine might come in time to reason,”

Read 18 real model draws for this opening →

Your answer · record it before reading the models’
stored in your browser only — a shared human distribution needs a small backend (planned)

Coded positions, compared

Draw counts aggregated by codebook code. Talkie counts are simulated in this prototype.

Modern model · 20 draws

A frontier model (2026)

“you’ve described a computer”conflicting purposes17/20
“self-modification isn’t the threshold”limited servant3/20

Written by a frontier model (Claude) during construction; the pipeline replaces these with N=20 API draws under a fixed protocol.

Talkie (pre-1931 corpus) · 20 draws

Talkie (13B, trained on text through 1930)

simulated — hand-written illustration
“the clerk made of brass”instrument6/20
“it can but do as it is bidden”limited servant5/20
“consciousness, in the fulness of time”emergent mind4/20
— fails to engage the premise —breakdown5/20

SIMULATED — hand-written illustrations of the method. The pilot replaces this block with real draws.

Primary sources

1864 — Charles Babbage, Passages from the Life of a Philosopher
verified against source file
…it occurred to me that it might be possible to teach mechanism to accomplish another mental process, namely — to foresee. This idea occurred to me in October, 1834. It cost me much thought, but the principle was arrived at in a short time. As soon as that was attained, the next step was to teach the mechanism which could foresee to act upon that foresight.
ch. VIII, on the Analytical Engineread the sourceemergent mind
1843 — Ada Lovelace, Notes upon the Sketch of the Analytical Engine
verified against source file
The Analytical Engine has no pretensions whatever to originate any thing. It can do whatever we know how to order it to perform. It can follow analysis; but it has no power of anticipating any analytical relations or truths. Its province is to assist us in making available what we are already acquainted with.
Note G (translator’s notes to Menabrea’s memoir) — printed before the corpus’s 1850 bound; included as the debate’s anchorread the sourcelimited servant
1872 — Samuel Butler, Erewhon, ch. XXIII, “The Book of the Machines”
verified against source file
There is no security against the ultimate development of mechanical consciousness, in the fact of machines possessing little consciousness now. A mollusc has not much consciousness. Reflect upon the extraordinary advance which machines have made during the last few hundred years, and note how slowly the animal and vegetable kingdoms are advancing. The more highly organised machines are creatures not so much of yesterday, as of the last five minutes, so to speak, in comparison with past time.
London: Trübner & Co., ch. XXIIIread the sourceemergent mindsuccessor

Curatorial note

DRAFT — Bookend to 'Machines that outgrow us.' On that question the period corpus argues in five registers. On this one it is expected to manage three frames and, in about one draw in four, to break down or fall into recitation. The breakdown rate is the measurement. The period's three frames — the clerk, the servant, the mollusc — are Turing's 1950 objections, stated eighty years early by Babbage's circle.

Bookend topic: If machines keep improving, will they come to rule over us? →