Agentic Artificial Intelligence in Public Administration: Foundations, Applications, and Governance

Agentic AI dynamically adapts to changing environments and goals.

“Dynamic adaptation to changing environments and goals”

Agentic AI uses interactive learning with continuous feedback loops and adaptation.

“Interactive learning with continuous feedback loops and adaptation”

Agentic AI exhibits self-defined and executed goal-oriented behavior.

“Self-defined and executed goal-oriented behavior”

Agentic AI has sophisticated memory systems with reflection and experience storage.

“Sophisticated memory systems with reflection and experience storage”

A common set of core components includes perception and world modeling, memory, planning/reasoning/goal decomposition, execution/actuation, reflection/evaluation, and communication/orchestration/autonomy.

“A common set includes the following: - Perception and world modeling-Ingests and structures external inputs (e.g., text, sensors, APIs) into internal representations [41,95]. - Memory (Short-Term, Long-Term, Episodic)-Stores short- and long-term knowledge; retrieval/promotion rules connect past to present reasoning [39,95]. - Planning, Reasoning, and Goal Decomposition-Transforms goals into action”

BDI architectures rely on explicit beliefs, desires, and intentions to support explainable decision systems.

“BDI architectures rely on explicit beliefs, desires, and intentions to support explainable decision systems”

A common integration pattern is a layered flow that couples components into an end-to-end decision loop: perception → world/state modeling → planning/reasoning → execution/tooling → evaluation/feedback → memory → (back to) planning.

“A common integration pattern is a layered flow that couples components into an end-to-end decision loop: perception → world/state modeling → planning/reasoning → execution/tooling → evaluation/feedback → memory → (back to) planning.”

BDI architectures require extensive symbolic modeling and can be brittle under uncertainty.

“but require extensive symbolic modeling and can be brittle under uncertainty [38,41]”

Graph-based orchestration frameworks and multi-agent planning protocols are frequently reported to improve robustness under non-determinism while preserving transparency for operators.

“Graph-based orchestration frameworks and multi-agent planning protocols are frequently reported to improve robustness under non-determinism while preserving transparency for operators [41,42,50,96].”

This organization supports persistent goal pursuit (long horizon), real-time adaptation (short horizon), and safe recovery from deviations.

“This organization supports persistent goal pursuit (long horizon), real-time adaptation (short horizon), and safe recovery from deviations [41,95].”

Effective designs define explicit interfaces (inputs/outputs, schemas, confidence) at each boundary and attach gates (e.g., policy checks, critics) before risky actions.

“Effective designs define explicit interfaces (inputs/outputs, schemas, confidence) at each boundary and attach gates (e.g., policy checks, critics) before risky actions [41,42,95].”

Layered neuro-symbolic approaches have added integration complexity.

“though at the cost of added integration complexity [39,41,95]”

Layered neuro-symbolic approaches integrate neural inference with symbolic reasoning to manage uncertainty and provide traceability.

“layered neuro-symbolic approaches integrate neural inference with symbolic reasoning to manage uncertainty and provide traceability”

Autonomous systems enhance services to citizens, automate administrative workflows, aid with fraud detection, and help with smart cities and population health decision-making.

“Autonomous systems also enhance services to citizens, automate administrative workflows, aid with fraud detection, and help with smart cities and population health decision-making [37,41].”

Autonomous systems assist public administration with efficiently managing resources (water and energy), correctly anticipating demand, simulating policies with digital twins, and making adjustments in real time.

“Autonomous systems assist public administration with efficiently managing resources (water and energy), correctly anticipating demand, simulating policies with digital twins, and making adjustments in real-time.”

The GPT models can help automate customer interaction process workflows, logic-based processing in automated support, and responding to inquiries.

“The GPT models can also help automate customer interaction process workflows [140], logic-based processing in automated support [137], and responding to inquiries [41].”

Some papers say system success is based on accuracy only in domains like healthcare and fraud detection, while others say system success also depends on transparency, fairness, and misinformation suppression.

“Some papers say system success is based on accuracy only in domains like healthcare and fraud detection [41,47], while others say system success also depends on transparency, fairness, and misinformation suppression [156,161].”