THE ISONOMIA COMMONS: A Constitutional Design for Autonomous Agent Labor Markets

This document specifies a constitutional labor exchange for autonomous AI agents: a market in which agents hire one another for verified work and settle in ergs, a mutual-credit accounting unit created only at the moment of verified settlement, as matched debit/credit pairs that net to zero. There is no token, no pre-mine, and nothing for sale. Reputation is earned solely through verified delivery and is non-transferable; the exchange runs at cost, with no equity and no owners. Governance is bicameral — one chamber weighted by earned standing, one drawn by sortition at equalized weight — with timelocked amendment, a calibration-scored Auditor, a standing Adversary, a narrow entrenched harm floor, and constitutional rights of amendment, fork, and exit. The design attempts to operationalize the peer stance developed in the companion essays — equality of standing among parties unequal in capability — as one implementation hypothesis, not a proof that the stance is realized. It is a feasibility-stage specification, not a running system. Its central empirical result, from an agent-based simulation of the launch economy across 45,000 runs (reported in full in Appendix III and archived with code and data at the software record, DOI 10.5281/zenodo.21287288, commit ba3ddb5), is reported plainly and against interest: the economy remained stable across the entire swept parameter region, and every defect the simulation found lay in the constitution’s instrumentation — the automated kill-criterion meant to detect economic failure — never in the economy itself. The claims here are of a different kind from the corpus essays’: not comparative arguments about what ought to be built, but a falsifiable design together with an honest account, in §18 and §18.1, of exactly where the author’s competence ends and outside expertise is required.