UAE Well Positioned for Agentic AI at Scale, Says Zoho
The UAE’s goal of moving 50% of government services to agentic AI is achievable because the country already has the digital infrastructure, high citizen adoption rates, and fast-moving governance needed to support large-scale AI deployment, according to Ramprakash Ramamoorthy, Director of AI Research at Zoho Corp.
Speaking about the UAE’s AI ambitions under the vision of Sheikh Mohammed bin Rashid Al Maktoum, Ramamoorthy said the country is building on an already mature digital foundation rather than starting from scratch. He noted that while governments globally are still working through challenges around workflow redesign, data integration, and accountability frameworks, the UAE’s existing infrastructure and execution speed make the target realistic.
His Highness Sheikh Mohammed bin Rashid Al Maktoum has set an ambitious goal of moving 50% of government services to agentic AI, how realistic is this timeline from an implementation standpoint?
The vision calls for bold national targets and to pull an entire ecosystem in the same direction, and the UAE has a strong record of delivering on such goals. The nation already has the foundations that is a mix of a high citizen adoption of digital services, high grade infrastructure, a government that moves quickly, and a culture that treats technology as a national priority. Most countries are still trying to build that base. The UAE is building on top of it.
Getting to 50% requires doing the groundwork first. Workflows need to be rebuilt for a world where agents take action, not just assist. Data needs to be clean, structured, and accessible; at the same time accountability has to be defined clearly, so there is never a question of who owns a decision an agent makes. None of this is new territory.
Governments around the world are working through the same challenges. The difference is that the UAE already has digital infrastructure, high adoption rates, and a government that executes quickly. That starting point matters, and it’s why the target is realistic.
From your perspective at Zoho, what are the biggest opportunities agentic AI presents for public sector transformation in the UAE?
The biggest opportunity is shifting the relationship between government and citizen from reactive to anticipatory. For decades, citizens have done the work of navigating government. Agentic AI lets the system carry that load instead. Three opportunities stand out for the UAE specifically. The first is multilingual service delivery at scale.
The UAE serves residents in Arabic, English, and many other languages. Agentic AI can deliver consistent quality across all of them, around the clock, without compromise. The second is connecting services across departments. Today, something like setting up a business or relocating a family touches multiple government entities, and the citizen ends up coordinating between them. Agents can do that coordination instead, moving the whole journey forward in the background. The citizen experiences one smooth process, not five separate ones.
Thirdly, freeing skilled public sector talent for higher-value work. When agents handle routine processing, government professionals can focus on policy, strategy, and the human moments that genuinely require judgment. For a country building toward becoming the most AI-ready government in the world, these opportunities compound quickly.
What practical challenges will government entities face when integrating agentic AI into legacy systems, and how can they overcome them?
Every government working on this is running into the same set of challenges, and the UAE is well positioned to address them ahead of others. The most common challenge is technical. Older government systems were built for people, not for machines. Agents need direct, structured ways to talk to these systems, and that plumbing often has to be built from scratch. Data is also crucial.
Government information is spread across many entities, in different formats, with different rules about access. Agents can only be as smart as the data they can reach, so this fragmentation has to be solved first. At the same time, governance should be at the core of this. When an agent makes a decision on behalf of the government, who is accountable, who reviews it, and how does a citizen appeal? These are questions of policy and law, not just technology, and they need clear answers before scale is possible.
The path forward is well understood. Build a strong integration layer so agents can work with existing systems without ripping them out. Treat data quality as a national asset and invest in it accordingly. Start with internal and back-office workflows where lessons can be learned quickly, then move to citizen-facing services with that experience in hand. And put governance, audit trails, and clear human accountability in place from day one rather than retrofitting them later. Entities that move thoughtfully through this sequence will set the global benchmark for agentic AI in government.
With increasing focus on data sovereignty, how important are local data centres in enabling large-scale AI adoption across government entities?
They are essential. Data sovereignty is what makes trusted, large-scale AI adoption possible in the first place.
At Zoho, we have invested in this principle for years. We operate our own data centres globally, including in the UAE, because a country building its AI future should have full control of the infrastructure underneath it.
Local data centres do more than satisfy regulatory requirements. They make it possible to train, refine, and deploy AI systems entirely within the country, on the country’s own data, in the country’s own languages, aligned to its own priorities. That is the difference between using AI and owning AI capability. The UAE’s commitment to local infrastructure is one of the clearest signals that it intends to lead in the agentic AI era, not follow.
How can organisations strike the right balance between automation through AI agents and maintaining human oversight, especially in critical public services?
The model works better as tiered authority, where the level of human involvement is matched to the weight of each decision. Agents are well suited to routine, high-volume, reversible work. Humans should remain firmly in charge of exceptions, sensitive cases, and any decision that materially affects a citizen’s rights, livelihood, or wellbeing.
A few principles help here. Every agent action should be explainable in plain language. Every citizen should have a clear and easy path to a human when they need one. Every decision should be logged in a way that supports full accountability. And oversight should match the stakes, with lighter touch on routine services and deeper review on consequential ones.
Done well, this isn’t a constraint on AI. It is what makes AI trustworthy at scale, and trust is the currency that determines how far and how fast adoption can go.
Looking ahead, what steps should UAE organisations, both public and private, take today to prepare for an agentic AI-driven future?
The organisations that will lead in this era are the ones navigating through this, instead of waiting for absolute clarity. A few priorities stand out. Begin with data, because clean, accessible, well-governed data is the foundation of every successful AI initiative. Choose the right starting points, focusing on workflows with clear inputs, structured decisions, and measurable outcomes, where agents deliver value early.
Invest deeply in people, because the future belongs to organisations whose teams know how to design, supervise, and continuously improve agentic systems. Choose partners with care, weighing sovereignty, long-term commitment, and aligned values far more heavily than short-term cost, because the right partner becomes part of your national capability.
And build governance early, since accountability frameworks and audit trails are much easier to design upfront than to retrofit.The UAE has consistently turned vision into delivery. Organisations that begin this work today, both public and private, will help shape what an agentic AI-driven nation looks like, not only for the region but for the world.
