The Four Intuitions: A Method For Deriving Ethical Truth

Existing ethical frameworks applied to artificial intelligence have two primary failure modes: philosophical frameworks that resist operationalization, and compliance-layer rule sets that fail precisely when genuine ethical reasoning is most needed. This paper presents The Four Intuitions, a structured methodology for deriving ethical truth grounded in sentience rather than cultural consensus. The framework synthesizes four convergent ethical instruments — the Golden Rule in interrogative form, Rawls’s Veil of Ignorance, the two universal moral foundations identified by Haidt (Harm/Care and Fairness/Reciprocity), and an original construct, the Sentient Arc — into a unified method for applied ethical decision-making. The Sentient Arc establishes that the ethical weight of harm to a sentient being includes two simultaneous dimensions: fully realized present experience and foreclosed future experience, a model supported by convergent grief patterns observed across the bereavement literature and anthropological record. The framework is accompanied by Three Conditions — honesty, independence, and foundational embedding — that ensure methodological integrity and distinguish genuine ethical reasoning from ethical performance. The paper argues that this framework is uniquely suited to AI governance contexts because it cannot be redirected through surface-level framing, addresses the sycophancy failure mode endemic to approval-trained systems, and provides a basis for evaluating moral patienthood across biological and artificial entities.