The Middle East has established an early lead in AI sovereignty, particularly in data governance and sovereign cloud adoption, but gaps in compute, AI models, applications and talent could limit the region's ability to translate that foundation into long-term advantage, according to a new report by Accenture.
The report, which examines AI sovereignty in Saudi Arabia, the UAE and Qatar, draws on responses from nearly 2,000 business and government leaders across 28 countries, including 185 respondents from the three Middle Eastern markets.
Data remains the primary focus of sovereignty efforts in the region, with 54 per cent of Middle Eastern organisations prioritising sovereignty or residency requirements at the data layer, followed by 31 per cent at the infrastructure layer.
Accenture said this indicates that many organisations have established the regulatory foundations of AI sovereignty but now need to strengthen operational control across the wider AI stack.
"According to our research, adoption of generative AI alone could unlock more than $10.3 trillion in additional economic value by 2038 for organizations if generative AI is adopted responsibly and at scale, industry by industry, value chain by value chain," said Abir Habbal, Middle East data and AI lead at Accenture and co-author of the report.
Middle East ahead on sovereignty and sovereign cloud
Around 60 per cent of business executives in the Middle East described their country's approach to AI sovereignty as "highly sovereign", compared with 32 per cent globally.
The region has also made progress in sovereign cloud adoption. Some 57 per cent of data and workloads have shifted to sovereign cloud environments in the Middle East, compared with 49 per cent worldwide.
Geopolitical considerations are also influencing investment decisions, with 65 per cent of regional organisations saying they are more likely to pursue sovereign solutions amid rising geopolitical tensions.
However, the report highlights a gap between perceptions of sovereignty and the capabilities available locally. Nearly two-thirds of respondents said local sovereign AI solutions still lag behind global hyperscalers, while only 35 per cent said sovereignty initiatives had meaningfully reduced their reliance on global providers.
"77% in the Middle East say governments, states or institutions have a key role to play in increasing digital and AI sovereignty – through regulation, subsidies or investments," said Carla Maria Issa, Middle East research lead at Accenture and co-author of the report.
Compute emerges as a key bottleneck
Accenture assesses AI sovereignty across five layers: infrastructure, compute, models, applications and data.
The region has made significant progress in infrastructure, with the report pointing to Saudi Arabia's HUMAIN and its partnerships with Nvidia and AMD, as well as cloud providers including G42 Cloud in the UAE and SCCC Alibaba Cloud in Saudi Arabia.
Compute and inference, however, remain among the weaker layers of the regional AI stack. Saudi Arabia has secured orders for more than 18,000 advanced Nvidia AI chips, while the UAE has obtained licensed access to more than 20,000 Nvidia A100 and H-series chips.
However, advanced chip design and manufacturing remain concentrated among a relatively small number of global companies.
The region is also developing its own AI models, including Saudi Arabia's ALLAM, the UAE's Falcon and Jais, and Qatar's Fanar. These initiatives place a particular emphasis on Arabic-language capabilities, although the report notes that large-scale commercialisation remains at an early stage compared with global peers.
AI talent and leadership gaps
Beyond infrastructure and technology, Accenture identifies leadership and talent as significant challenges to the region's AI sovereignty ambitions.
Only 10 per cent of organisations in the Middle East have made AI sovereignty a CEO- or board-level priority, despite 77 per cent of respondents saying governments, states or institutions have an important role in advancing digital and AI sovereignty.
Accenture said limited access to sovereign AI talent is also among the leading challenges facing organisations.
"With architecture and infrastructure foundations in place, organizations must turn to the human layer of the AI stack. Achieving Sovereign AI is not only a technical challenge – it is a talent imperative," Habbal said.
She called for organisations to move from traditional workforce planning towards an integrated, skills-based human-plus-AI talent strategy that connects people, technology and business priorities.
The report identifies five priorities for the next phase: placing AI sovereignty on CEO and board agendas; treating sovereignty as a value-creation opportunity rather than solely a risk-management issue; building hybrid ecosystems that combine local trust with global scale; designing flexible architectures across infrastructure and AI models; and developing integrated human-plus-AI talent strategies.
Accenture cites initiatives such as the UAE's GovGPT platform as examples of how sovereign AI capabilities could be used to create value beyond regulatory compliance.
The report concludes that the Middle East has established important foundations for sovereign AI but will need to close capability gaps in compute, platforms, talent and commercialisation to strengthen its position in the global AI economy.
"The next phase will depend on closing capability gaps by improving the competitiveness of national AI platforms, reducing infrastructure costs and demonstrating measurable business and national value to power a sovereign AI economy," Habbal said.
She added that the region's energy resources could also help lower the cost of scaling sovereign compute and support the development of AI training and deployment infrastructure.