How Agentic AI could reshape government services in Abu Dhabi

Agentic AI could change how governments deliver public services by moving beyond answering questions to completing tasks, coordinating across departments and responding proactively to people’s needs, according to experts from Deloitte Middle East.

Speaking about the growing role of the technology in government, Mike Horseman, Partner at Deloitte Middle East, said AI agents can already understand requests, gather information from different government systems and complete administrative processes.

“We are also seeing agents support procurement, case management, regulatory compliance, and internal operations by coordinating work across departments,” Horseman said.

He added that the main opportunity is not simply making existing services faster, but redesigning them around people’s needs.

As governments increase the use of AI, human oversight will remain important, particularly for decisions carrying greater risks.

Horseman said governments should give AI agents only the level of independence needed for a particular task, while higher-risk decisions should remain under meaningful human oversight.

Strong cybersecurity, identity management, data sovereignty and continuous monitoring should also be built into AI systems from the start.

“Every action taken by an AI agent” should be explainable and auditable, he said, helping governments innovate while maintaining public trust.

Shargil Ahmad, Partner and Government and Public Services Leader at Deloitte Middle East, said government officials will continue to set policy, design services and remain accountable for how government operates.

He identified critical thinking, consultation, AI orchestration, human-centred design, change leadership and decision-making among the skills that will become increasingly important for government employees.

One of the biggest challenges is moving Agentic AI beyond individual pilot projects.

Horseman said successful implementation requires connected data, secure integration and modern digital platforms that allow AI agents to work across government systems.

Governance will also need to evolve as AI becomes capable of acting independently.

Governments will need clear rules defining what an AI agent can do, when human approval is required and how its decisions are recorded.

Ahmad said Abu Dhabi is well positioned to become a leading AI-native government because of its growing AI talent, digital infrastructure and clear policy direction.

“Together with the leadership’s coordination of government entities, ongoing investments, and our culture of responsible innovation, we are uniquely well-positioned to become the world’s first AI-native government,” he said.

At the DGE Showcase, Deloitte presented AI Core alongside Agentic AI use cases built on the Common Digital Platform architecture supporting Abu Dhabi Government.

The experts said a common digital foundation could help Abu Dhabi move from individual AI experiments towards wider implementation across government, while maintaining security, governance and interoperability.