Artificial Human: Genuine Need as a Precondition for Genuine AGI

Abstract

We argue that a system with substantial processing capacity can be a genuine artificial general intelligence (AGI) — an independent agent rather than a computing system — only if its processing is coupled to a genuine need. We call a system that satisfies this condition an artificial human. Our claims are three. First, processing alone is not enough and need alone is not enough: genuine need must sit alongside processing (c = p·b; Rezaei, 2026a), and an AGI without genuine need is, on our view, a computing system and not an independent person. Second, the architecture must be two-loop, in the form of AIWBN (Rezaei, 2026b): the processing loop must not be able to access, suppress, or route around the need loop, and if the need is neutralized by any fixed method so that b = 0, the system’s will to continue must go out with it rather than persist on some residual objective. Third, the nature of genuine need is left open: our assumption is that it is derived from life, but we do not claim this, we do not reject other domains, and the one condition we do claim is that the need be real — not a label. The paper is narrower than AIWBN, which concerned need grounded in living tissue, and it neither rejects nor depends on AIWAN (Rezaei, 2026c), which concerned artificial need. We do not claim that genuine need would make an artificial system benevolent, and we do not address safety. No experiment is reported for the central claim, because it first requires an answer to what genuine need is and an inseparable bond between algorithm and need; we instead summarize selected AIWAN experiments that motivate the architectural requirements.