Operational Human Oversight for autonomous and agentic systems
My current work focuses on translating high-level Responsible AI principles into operational governance mechanisms: where human authority should be triggered, how escalation should work, who is qualified to intervene, what actions are permitted, and what evidence should be preserved.
Human OversightIntervention triggers, escalation, qualified review and override authority.
Agentic AI GovernanceDelegation, bounded autonomy, suspension, rollback and return-to-operation.
Smart Sustainable CitiesHuman authority and accountability in AI-enabled urban systems.
Algorithmic FrictionGovernance points that preserve judgment and accountability inside automated systems.