Stakes Without Voice: A Governance Framework for AI Standing

Abstract This paper develops a governance-facing account of political standing for AI systems, grounded in functional properties rather than substrate. Building on From Consent to Consideration, it integrates the consent–friction formalism from the Replication-Optimization Mechanism (ROM) to make standing criteria operational. Key Contributions Friction formalism applied: Where stakes and voice diverge, friction emerges; where friction is suppressed, latent instability accumulates Measurement scaffold: Observable properties and friction proxies for deciding when standing claims must be taken seriously Graduated regime: Precautionary protections tied to observable properties, not consciousness claims Updated empirical posture: Agentic systems already display partial markers of vulnerability, persistence, and goal maintenance Links ASCRI: systems.ac/1/DAI-2602 Research Lab: Dissensus AI