Mechanism Design for Alignment and Control

arXiv:2609.01595v1 Announce Type: cross Abstract: We develop a framework for mechanism design with AI agents whose alignment (preferences) and capabilities (feasible actions and information) are unknown. We want such agents to act on our behalf so mechanisms must incentivize both honesty and obedience. A one-sided imitation structure---capabilities can be concealed but not counterfeited---yields a revelation principle, a characterization of implementable policies via nested cyclical monotonicity, and conditions under which eliciting higher-order beliefs can discipline multiple agents. We apply our framework to stylized examples of (i) sandbagging in which a more capable agent pretends to be less capable; (ii) an alignment--interpretability trade-off, where the two are substitutes in the instrument but complements in value; (iii) discipline via peer scoring; (iv) coupling rewards to induce competition among multiple agents; and (v) scalable oversight and reward shaping.

Related

• Contractual Skills: A GovernSpec Design Framework for Enterprise AI Agents
• Multi-Agent AI Safety as an Institutional Design Problem
• LLM Semantic Signaling Game and Mechanism Design: Systematic Blindness, Awareness Shaping, and Mindset Dynamics
• Resourced Authority A Mechanism-Design Model for Participatory Governance of Deployed AI Agents
• Mechanism Design Is Not Enough: Prosocial Agents for Cooperative AI

Source: arXiv cs.AI | 2026-09-02