The United Arab Emirates (UAE) is a pioneer in the integration of artificial intelligence (AI) within governmental frameworks, having launched its national AI strategy back in October 2017. This initiative included appointing Omar Sultan Al Olama as the world’s first Minister of State for AI at just 27 years old. The UAE has made remarkable strides in developing digital identities, sovereign cloud frameworks, and data-sharing protocols that many nations are still in the process of establishing. The success of agentic AI in government operations is now a critical focal point to assess whether this early investment will yield substantial benefits.
Initiation of the Agentic AI Project
This month marked a significant milestone as over 100 federal officials gathered for a workshop designed to kick off the UAE’s agentic AI project. The goal is ambitious: to transition half of federal government operations and services to agentic AI models within just two years. The National Committee for the Agentic AI Project orchestrated this initiative, setting clear priorities and timelines while beginning the groundwork to determine which governmental tasks can effectively be automated.
A crucial aspect of this classification process will dictate the operational landscape. While the UAE boasts advanced computational capabilities—including an AI-driven performance system that tracks over 150 million data points each month—no government has yet established a clear framework delineating which functions can be completely automated versus those that should remain human-driven.
Governance and Human-AI Collaboration
Huda Al Hashimi, Deputy Minister of Cabinet Affairs for Strategic Affairs, noted that the recent workshop signifies the practical implementation phase of AI models across governmental activities. This expansive program is structured around seven foundational pillars: strategy and projects, foresight and intelligence, policy and governance, government performance, competitiveness, and innovation. Furthermore, discussions at the workshop also addressed performance indicators that will guide the initiative while ensuring different entities avoid redundant efforts.
A guiding principle drives this project: “humans lead, AI enables.” When announcing the strategy, Sheikh Mohammed bin Rashid Al Maktoum emphasized the UAE’s ambition to become the first government globally to deploy agentic AI extensively across its sectors. This balance—between autonomous functionality and human oversight—highlights the complexity of AI integration in public services. Until the classification frameworks are finalized, questions surrounding accountability for automated decision-making remain unanswered.
The Significance of Scale
Sheikh Mansour bin Zayed Al Nahyan oversees this significant initiative, supported by a task force chaired by Mohammad Al Gergawi, the Minister of Cabinet Affairs. Recent developments have seen the Cabinet approve a suite of government services to incorporate agentic AI tools, alongside a historic training program that encompasses 80,000 federal employees. Additionally, the Cabinet approved a national AI healthcare policy, solidifying the digital record as the official repository for core government data, fostering secure information sharing across different agencies.
The timeline for these initiatives has been carefully planned. A recent training session in Dubai saw participation from over 300 representatives from 50 federal entities, with aims to identify suitable services for AI integration within a tight 90-day window. Another workshop conducted by the Presidential Court and the Ministry of Cabinet Affairs involved training 600 employees using similar content.
Global Implications of Agentic AI
The move toward agentic AI in government takes on a more expansive relevance beyond the UAE. Countries in both Asia and Europe are starting their journeys into AI adoption, often with lesser-defined strategies and no concrete timelines. By establishing clear objectives and deadlines, the UAE is positioning itself as a global leader in demonstrating what agentic AI in government can accomplish. Within two years, we will have empirical data to evaluate how this technology impacts service delivery, staffing, and error rates.
The classification frameworks that will be developed during this journey are particularly crucial. Their release will represent the first meaningful attempt by any government to formalize the decisions it is comfortable allowing machines to autonomously execute. Everything else in this ambitious program—training, procurement—depends on establishing these foundational guidelines, which will be paramount for a future where AI plays a critical role in governmental functions.