Earlier this year, the UAE government announced one of the world’s boldest AI commitments. Within two years, half of all federal government operations will run on agentic AI. Fifty government entities have each been tasked with taking at least one AI-powered service from concept to live deployment within 90 days. It is an ambitious target, but perhaps not as surprising as it first appears. It reflects a pattern taking shape across the wider Middle East region.

In both the UAE and Saudi Arabia, research shows organisations already rank among the world’s most advanced adopters of agentic AI. More importantly, public and private sectors in both markets are clearly moving in the same direction. Governments are embedding AI into public services while enterprises are accelerating deployment across their own operations.

That convergence suggests something more significant than widespread AI adoption. The region is effectively addressing the underlying infrastructure necessary to make AI work reliably at scale, a layer many markets have yet to confront.

The challenge is no longer AI

For the past few years, the AI conversation has centred on the models themselves, how capable they are and which use cases they unlock. Agentic AI changes that. Unlike systems that generate content or offer recommendations, agentic AI makes decisions and takes actions autonomously, which means every decision depends on accurate, real-time information rather than yesterday’s data or an overnight batch update. That shifts the emphasis from the model to the data feeding it, and specifically to real-time streaming, the continuous movement of information as events happen that gives applications, systems and AI models a shared, live view of the business. For autonomous AI, that is now an architectural imperative.

This shift already shows up in investment priorities. Globally, organisations place similar strategic importance on AI and data streaming. In the UAE and Saudi Arabia, streaming pulls markedly ahead, with around nine in ten IT leaders naming it a strategic priority. That reflects a region that has already recognised that the biggest barrier to scaling AI is less the intelligence of the models than the quality, availability and movement of the data behind them.

That readiness didn’t happen overnight. For more than a decade, Gulf governments have invested in cloud adoption, digital government, smart cities and connected infrastructure. While these programmes were never designed as AI initiatives, they quietly built the real-time, connected foundations agentic AI now depends on.

That early, patient investment is now becoming a competitive advantage. Many organisations across the region were able to build modern platforms around live, connected data from the outset, rather than retrofitting architecture built for a different era. That is exactly the foundation operational AI requires.

Operational AI demands operational data

To be sure, this does not mean that the difficult work is finished. Two-thirds of IT leaders in both the UAE and Saudi Arabia still identify data quality and infrastructure as major obstacles to scaling agentic AI. If anything, that is reassuring. It suggests organisations understand exactly where the real challenges lie.

Pilots can tolerate incomplete data, manual intervention and the odd error. Operational AI cannot. Once autonomous systems start making decisions that affect customers, employees or public services, trust becomes inseparable from the quality of the data behind them. Reliable infrastructure stops being a technical concern and becomes a business requirement.

Encouragingly, the region appears to be responding. Almost all IT leaders in both countries believe improving real-time data capabilities will accelerate agentic AI adoption and increase returns on AI investment. That is evidence that infrastructure is no longer something to fix after AI is deployed. It is becoming part of the AI strategy itself.

Setting the global blueprint

Initiatives and investment across the region, including the UAE government’s agentic AI drive, reflect that same evolution. This is not just about encouraging AI experimentation, but about building the conditions for AI to become embedded in the everyday delivery of public services. Enterprises across the Middle East are on a similar path, moving beyond pilots toward operational deployment.

The region’s lead in agentic AI is what has put it in the global spotlight. But the more important signal is what happens next. For AI to keep accelerating, it needs data streaming to lead from the front rather than trail behind as an afterthought. The Middle East has shown maturity on both fronts, moving fast on adoption while building the infrastructure to sustain it, and the next two years will be worth watching closely. If it gets this right, the Gulf will not just have adopted AI early. It will have written the playbook the rest of the world follows.

By Karim Azar – AVP & GM, Confluent Middle East

Related post

View all