The Gulf faces a digital sovereignty dilemma marked by strong infrastructure but gaps in computing and executive leadership

A report from Accenture shows that the Middle East has made early progress in AI governance and sovereign cloud adoption, but it now faces a critical juncture that requires deeper investment in computing layers, model and application development to turn this initial progress into a sustainable competitive advantage. The report is based on a survey of about two thousand leaders in the public and commercial sectors across 28 countries, including a focused sample of 185 participants from Saudi Arabia, the United Arab Emirates and Qatar, to map the current state of digital sovereignty and its maturity paths in the region.

The survey results highlighted a strong focus on data protection, with 54 % of regional organizations prioritizing data localization and on-premise storage first, followed by the infrastructure layer at 31 %. Abir Habbal, head of data and AI at Accenture Middle East and a contributor to the report, notes that generative AI alone could generate additional economic value exceeding $10.3 trillion by 2038 if deployed responsibly and at scale across sectors and value chains.

Regulatory confidence in the region collides with an operational capability gap and continued reliance on global providers.

The report reveals a striking paradox between confidence levels and actual capabilities, with 60 % of Middle Eastern CEOs describing their countries’ approach as \"high-sovereign\" compared with only about 32 % globally. Regional organizations have moved 57 % of their data and workloads to sovereign cloud environments versus 49 % worldwide, and 65 % say they are inclined to adopt sovereign solutions amid rising geopolitical tensions. Despite these positive signals, nearly two-thirds of respondents believe local sovereign solutions still lag behind global cloud giants, while only 35 % acknowledge that current efforts have materially reduced their operational reliance on international providers. Carla Maria Issa, head of Middle East research at Accenture and a contributor to the report, points out that 77 % of participants rely on governments’ pivotal role in advancing sovereignty through legislation, support and direct investment.

The report analyses sovereignty across five integrated technology layers, starting with infrastructure, then computing, models, and finally applications and data. The region’s strength is evident in infrastructure, thanks to major projects such as the HUMAIN company in Saudi Arabia and its partnerships with Nvidia and AMD, alongside cloud platforms like G42 Cloud in the UAE and Alibaba Cloud in Saudi Arabia. However, the report identifies the computing and inference layer as the most prominent bottleneck; although Saudi Arabia has ordered more than 18,000 advanced AI chips from Nvidia and the UAE has obtained licences for over 20,000 A100 and H-series chips, the manufacturing and design of advanced chips remain confined to a few global firms. In the domain of language models, the region continues to develop pioneering Arab initiatives such as Saudi Arabia’s \"Alam\" model, the UAE’s \"Falcon\" and \"Jais\", and Qatar’s \"Fanar\", yet the large-scale commercial rollout of these models is still in its early stages compared with the global landscape.

Limiting sovereignty to legal compliance corridors deprives companies of building a genuine competitive advantage in the market.

The most alarming finding is the leadership gap, with data showing that only 10 % of Middle Eastern organizations have placed AI sovereignty on the agenda of their boards or CEOs. Accenture regards this shortfall as a critical blind spot, warning that the absence of executive adoption reduces sovereignty to a merely formal regulatory compliance measure rather than leveraging it as an engine of economic value, and that this is accompanied by a scarcity of specialized talent to lead the ecosystem locally and manage the integration of human capabilities with technology.

For technical and executive leaders in the Gulf, Egypt and the Levant, these results require an immediate reshuffling of investment priorities. It is no longer sufficient to lease local storage space or merely comply with data residency laws on nearby servers; organizations must build a flexible architecture that does not rely on a single cloud provider and focus on the operating cost of inference workloads and specialized application development. Moreover, the Gulf’s relative advantage in providing secure, low-cost energy supplies gives firms the opportunity to reduce the expense of computational expansion, enabling regional companies to train and deploy their models locally at competitive cost and equip engineering teams capable of managing and adapting sovereign models instead of consuming off-the-shelf APIs.