Why sovereign and agentic AI will define next phase of Middle East’s digital transformation
Organisations shift from artificial intelligence pilots to operational deployment as governments prioritise digital sovereignty and infrastructure control
Digital transformation in the Middle East is reaching a critical turning point. After years of rapid experimentation with artificial intelligence (AI), cloud platforms and automation, organisations are now being forced to answer a harder question: how to scale advanced technologies in a way that delivers measurable value, preserves trust and strengthens national digital sovereignty.
According to Cliff de Wit, group chief innovation officer at Accelera Digital Group (ADG), the region is approaching a decisive moment. “What we are seeing now is a clear shift away from isolated pilots towards AI becoming part of core operational infrastructure,” he says. “This year, the organisations that succeed will be those that treat AI as a foundational capability rather than an innovation experiment.”
One of the most significant shifts underway is the move from traditional automation towards agentic artificial intelligence. Unlike earlier generations of AI, which were largely reactive or task-specific, agentic systems are designed to plan, execute and optimise workflows with minimal human intervention.
“This is about moving beyond chatbots and scripted automation,” says de Wit. “AI agents are now able to coordinate complex tasks, learn from outcomes and continuously optimise performance. That changes productivity, service delivery and even how organisations structure their teams.”
For governments and enterprises across the Gulf, this transition is particularly attractive. Large, complex organisations with extensive operational footprints stand to gain disproportionately from systems that can operate at scale, across multiple functions and geographies. But it also raises new questions around governance, accountability and skills.
Sovereign AI becomes a strategic priority
Alongside the rise of agentic systems, the Middle East is intensifying its investment in sovereign AI and compute infrastructure. Hyperscale datacentres, national cloud platforms and country-specific AI models are increasingly viewed as strategic assets rather than purely commercial investments.
“Beyond innovation, this is about digital independence,” explains de Wit. “Control over data, compute and AI capability is becoming a matter of national resilience and long-term competitiveness.”
This push reflects a broader geopolitical reality. As global supply chains fragment and technology becomes more closely tied to national interests, Middle Eastern governments are seeking to ensure that critical digital capabilities remain under local control. The result is a surge in investment not only in infrastructure, but also in domestic AI talent and locally trained models.
“We have already seen what happens when organisations rush to replace humans outright. Service quality drops, trust erodes and the technology ends up creating more problems than it solves. AI can scale processes, but only people can scale trust, creativity and culture”
Cliff de Wit, Accelera Digital Group
The region’s much-publicised smart city initiatives are also evolving. Early projects focused on digital services and connectivity. Today, cities are beginning to function as integrated, AI-driven systems that continuously optimise the physical environment.
“We are entering the era of physical AI. Mobility, energy, public safety and citizen services are being optimised in real time, at scale, using live data and AI-driven decision-making,” adds de Wit.
This shift has profound implications. As AI systems increasingly control physical assets and infrastructure, reliability, security and human oversight become critical. The technology may be invisible to citizens, but its impact on daily life will be tangible.
Despite the region’s ambition, scaling AI remains a challenge. De Wit notes that while AI adoption in the Middle East is exceptionally high by global standards, many organisations struggle to move beyond proof-of-concept deployments. “The issue is not enthusiasm,” he says. “It’s fragmented data, unclear operating models and, ultimately, the need to demonstrate return on investment. Chief financial officers want evidence of measurable business impact, not just technical capability.”
Some sectors have already crossed that threshold. Energy companies such as Saudi Aramco and Adnoc now run AI at an industrial scale, embedding it directly into production, maintenance and subsurface analysis. Industrial-scale language models trained on decades of proprietary data are enabling real-time decision-making, while predictive maintenance systems driven by vast sensor networks are delivering significant gains in efficiency and output.
Similar progress is visible in logistics and retail, where AI is being used to manage last-mile delivery, simulate disruption scenarios and optimise warehouse operations through digital twins. In several cases, organisations have reported substantial improvements in inventory accuracy and operational resilience.
Why people still matter
As automation accelerates, de Wit is emphatic that human expertise remains indispensable. “We have already seen what happens when organisations rush to replace humans outright. Service quality drops, trust erodes and the technology ends up creating more problems than it solves,” he says.
The most effective strategies are to utilise AI to eliminate friction and repetitive tasks, while retaining humans’ responsibility for judgement, empathy and accountability. “AI can scale processes, but only people can scale trust, creativity and culture.”
The security implications of this shift are significant. AI-powered search tools are becoming powerful reconnaissance engines for attackers, while autonomous agents are increasingly used on both sides of the cyber battlefield. “We are moving towards an agent-versus-agent environment. Defensive AI will need to monitor, predict and counter offensive AI in real time,” he adds.
For Middle Eastern organisations, the next two years will be critical. The region’s willingness to invest, experiment and move quickly has created a strong foundation. The challenge now is to translate that momentum into scalable, secure and human-centred digital transformation.
“The winners in 2026 will be those who understand that AI is not just a technology shift, but an organisational, cultural and strategic one,” de Wit concludes.
