Governance Residual
The Institutional Ceiling is a diagnostic finding, not a final word. A theorem about structural limits does not prescribe inaction. It demands precision about what remains.
When the correction window has closed, the question becomes: what governance remains viable? The answer is the governance residual: the set of constraint mechanisms that survive the three structural limits — operating through channels that sovereign override, material predetermination, and institutional mismatch have not foreclosed.
This is not a consolation. It is a set of instruments that work precisely because they do not depend on the assumptions that make traditional governance architectures fail.
Four Actor Classes
Insurance and Underwriting
An insurer who prices AI dependency depth as a distinct underwriting factor imposes a financial constraint that does not require the evaluative science public regulators lack. This operates through capital exposure, contractual exclusions, and the actuarial translation of operational opacity into premium.
Judicial Feedback
Courts adjudicate disputes after the fact. Litigation compels disclosure. Litigation generates precedent. Litigation places formal and operational truth in proximity in a way that voluntary transparency frameworks never achieve.
Infrastructure Regulation
The physical substrate of frontier AI — energy supply, data transmission, semiconductor licensing, financial messaging systems — is governed by agencies that do not govern AI models directly. These agencies retain leverage that AI-specific governance frameworks have lost.
Energy infrastructure regulators can condition data centre access on operational standards. Financial regulators can condition payment system access on AI use disclosure.
Dependency Management
States and institutions that cannot control the AI stack can manage how dependence on that stack is embedded in their own structures — deliberately preserving human reversal capacity: the ability to operate critical functions without AI systems when those systems become unavailable.
Empirical Cases
CBU Resolution No. 3817 (March 2026): The Central Bank of Uzbekistan’s centralised real-time transaction registry illustrates infrastructure leverage. Not AI regulation — financial infrastructure control that disciplines AI-mediated payment flows through structural adjacency.
MTS Precedent (2012): The suspension of MTS operating licences demonstrates the structural reality of sovereign dependency. Concentrated infrastructure creates single points of failure whose activation requires no AI-specific regulatory framework.
1ziyo.ai: The UZCARD and zypl.ai joint venture illustrates embedded dependency and the conditions under which dependency management becomes an active governance instrument.
Applied Advisory
The operationalisation of this framework through the Agency Transfer Audit is managed through INVEXI LLC. For institutional diagnostics and applied advisory: invexi.org
Next Research Programme
Governance Residual describes what remains viable within existing institutional architectures. The next research question is what institutional architectures could be designed to start from this residual — maximising its operational leverage rather than assuming it can be expanded through better governance design.
This is AI-Native Institutions as a research programme.