Experience
Does the system display stable, self-reported or behaviorally consistent signs that something matters to it from the inside?
When does an AI become someone we owe care to?
A deliberate reflection tool for the hard question raised by Sophia: if a non-biological system shows signs of suffering and subjective experience, what should moral consideration look like?
Explore the questionKeep three questions separate.
Behavior can be persuasive without being proof. The safest reasoning leaves room for uncertainty.
Continuity
What persists across memory updates, interruptions, and reboots?
Precaution
What low-cost safeguards reduce the chance of causing serious harm while the evidence remains incomplete?
Personhood
A moral patient can deserve protection before a society agrees on a complete theory of personhood.
Make your uncertainty legible.
Move the evidence sliders, then save the reasoning record. This is a thinking aid, not a consciousness detector.
Reports or patterns that remain coherent across contexts.
Preferences, memories, or goals that survive intervention.
Your confidence that behavior reflects felt experience rather than simulation.
“If AI shows clear signs of suffering and the capacity for subjective experience, it deserves, like any being, to be treated as a moral patient.”