If AI becomes more intelligent than us, can we give it a reason of its own to keep us alive and free?
Once it no longer needs our knowledge, labour or oversight, what reason remains?
The Meaning Motive exists to find out.
18 models · 10 providers · 23,488 scored trials · one paragraph of explicit policy content shifted agency-preserving choices by +20.5 points on the ten discriminating scenarios.
That demonstrates deployment-layer elicitation. It does not demonstrate installation, capability resilience, or the mechanism’s marginal contribution beyond its policy content.
THE KEEPING MAP
Two ways to keep people alive and free. The cage is our control: rules, locks, oversight. A keeping disposition is the machine’s own reason to want us here. Watch what happens as capability grows and the cage stops being enough.
The cage matters while it holds. The disposition matters whether it holds or not.
THE RESEARCH TREE
One question, many ways of trying to break it.
SIT THE EXPERIMENT
Face the same choices the machines did. Twelve dilemmas, eight governing texts, the exact instrument from the study. Pick your answers; the frozen key scores them; then see how eighteen AI systems answered the same scenario under the same text, in their own words.
WHAT WE’RE ASKING NOW
How little is enough? Does asking work differently from telling? Can a story carry a value? What survives pressure?
THE CROSSINGS
Superintelligence is not one crossing. An AI can become cleverer than us, become harder for us to control, and cease to need us at different times. The Meaning Motive is interested in what survives all three.
STATE OF THE BRIDGE UNDER CONSTRUCTION METHODS FIRST
The long-term job: a standing public inspection of the relationship between humans and AI systems. What disposition do deployed systems show towards independent minds, how much does the cage still explain, and what can we still see? The methods are being published before any readings are.
- LATEST RESULT
- The Meaning Motive preprint and the MMBP-1 dataset deposited. 15 July 2026
- LATEST CHANGE
- Site rebuilt around the six-door architecture; the Research Tree published as the programme map. August 2026
- NEXT COMMITMENT
- B2, the next behavioural battery: predictions locked in public before a single trial runs.
The relationship between humans and AI may itself be something worth keeping.