Saudi Arabia’s push to modernize public services is increasingly moving beyond digitizing existing processes toward building AI-enabled government systems. At LEAP 2026 in Riyadh, Hong Kong’s Applied Science and Technology Research Institute (ASTRI) and Saudi Arabia’s Digital Government Authority (DGA) signed a memorandum of understanding that creates a framework for collaboration on artificial intelligence, emerging technologies and digital government innovation.
The next phase of digital government is unlikely to be defined simply by moving government forms and services online. Governments are increasingly experimenting with artificial intelligence, automation and data-driven systems that can change how public services are designed and delivered.
A new partnership between ASTRI and Saudi Arabia’s Digital Government Authority is aimed at that transition.
The two organizations signed an MoU at LEAP 2026 in Riyadh, establishing a framework for cooperation spanning AI research, emerging technologies, talent development and practical technology testing.
Rather than focusing on a single AI product, the agreement creates several channels through which researchers and government technology teams can collaborate, develop ideas and test them in real-world environments.
That structure could prove important as governments attempt to move AI projects beyond demonstrations and into production.
Four pillars for government AI collaboration
The partnership is built around four areas.
The first is knowledge exchange, including joint technical workshops covering AI and emerging technologies. The second focuses on talent development, with plans for researcher secondments, PhD exchanges and internships.
The third involves joint research in artificial intelligence, digital government and emerging technology. The fourth is proof-of-concept validation, using the DGA’s sandbox environment to test potential solutions.
The sandbox component is particularly significant.
Government agencies often face a gap between an AI system that works in a controlled demonstration and one that can safely operate within public-sector infrastructure. Testing technologies in a controlled environment allows organizations to assess technical performance, integration requirements and governance considerations before broader deployment.
For enterprise and government AI teams, that type of experimentation infrastructure can be as important as access to AI models themselves.
Why Saudi Arabia is becoming an AI testing ground
Saudi Arabia has made digital transformation a central component of Vision 2030, with government digitization and emerging technology development playing an increasingly prominent role in the country’s economic strategy.
The country is also investing heavily in AI infrastructure, cloud computing, data centers and digital public services.
That creates an environment where international research organizations can work with government institutions on applied technology rather than limiting collaboration to academic research.
ASTRI brings a different but complementary ecosystem.
Based in Hong Kong, the institute focuses on applied research and technology commercialization, working across areas including artificial intelligence, communications, cybersecurity, smart cities and advanced technologies.
The new partnership therefore connects Hong Kong’s applied research capabilities with Saudi Arabia’s government-led digital transformation agenda.
From research to deployable technology
One of the biggest challenges facing government AI programs is translating research into usable systems.
AI research can produce promising models and prototypes, but government deployment introduces requirements around security, data governance, interoperability, reliability and public-sector accountability.
The ASTRI-DGA framework addresses part of that problem by connecting research and talent programs with proof-of-concept testing.
That model resembles a broader trend across the technology industry.
Companies such as Microsoft, Google, Amazon and NVIDIA are increasingly working with governments and enterprises on AI infrastructure and deployment environments. At the same time, organizations are building dedicated sandboxes and innovation centers where emerging technologies can be tested before being introduced into critical systems.
The value of these environments is not simply technological.
They allow institutions to determine which AI applications are genuinely useful before committing significant resources to production deployment.
Talent is becoming a strategic AI asset
The partnership’s emphasis on researcher exchanges and internships also highlights another constraint on AI adoption: skills.
Governments can procure cloud infrastructure and AI software relatively quickly. Building teams capable of designing, evaluating and governing those systems is harder.
Researcher secondments and PhD exchange programs can create longer-term connections between academic institutions, government technology teams and industry.
For Saudi Arabia, that could help strengthen domestic capabilities as demand grows for AI researchers, engineers, data specialists and digital-government architects.
For Hong Kong, the partnership offers another avenue for its applied technology ecosystem to engage with one of the world’s fastest-growing digital transformation markets.
A model for cross-border applied AI
The partnership also reflects the increasingly international nature of AI development.
AI infrastructure may be global, but the applications governments build around it are highly local. Language, regulation, public-sector processes, cybersecurity requirements and cultural expectations all influence how systems need to operate.
That makes cross-border partnerships particularly valuable when they combine technical research with local implementation expertise.
For DGA, collaboration with an applied research institute provides access to additional technical capabilities and research networks.
For ASTRI, the agreement creates a pathway into the Middle East while providing opportunities to validate technologies against large-scale government use cases.
The real test will be what reaches production
An MoU is not itself evidence that a particular AI system has been successfully deployed.
The more meaningful measure will be whether the collaboration produces working prototypes, validated technologies and eventually production services.
That makes the planned sandbox activity worth watching.
If the partnership can move projects from joint research into validated government applications, it could provide a practical model for how research institutions and digital-government authorities can collaborate on AI.
The timing is also notable. As governments worldwide move from generative AI experimentation toward agentic AI, intelligent automation and AI-enabled public services, the need for controlled testing environments is increasing.
Saudi Arabia’s collaboration with ASTRI places that experimentation within the country’s broader digital transformation strategy.
For the global government technology sector, the bigger story is the emergence of cross-border AI partnerships designed not merely to study emerging technology, but to test whether it can work in the real world.
Market Landscape
Government AI adoption is moving through several stages: experimentation, controlled pilots, production deployment and eventually integration into core public services.
The ASTRI-DGA partnership sits primarily between research and validation, an important stage where governments can assess technologies before making larger investments.
| Area | Traditional approach | Emerging AI-government model |
|---|---|---|
| Research | Academic or government-led | Cross-border research partnerships |
| Testing | Small internal pilots | Dedicated AI sandboxes |
| Talent | Permanent recruitment | Exchanges, secondments and research programs |
| Deployment | Large upfront programs | Proof-of-concept validation before scaling |
| AI strategy | Individual applications | Connected digital-government infrastructure |
The competitive landscape is also expanding beyond government IT vendors. Hyperscalers including Microsoft Azure, Google Cloud and Amazon Web Services, alongside AI infrastructure companies such as NVIDIA, are becoming increasingly important to national AI strategies.
For government technology leaders, the critical question is shifting from Can we use AI? to How do we test, govern and scale AI safely?
That makes applied research partnerships and controlled experimentation environments increasingly relevant.
Top Insights
- ASTRI and Saudi Arabia’s DGA established an AI research partnership, combining Hong Kong applied research expertise with Saudi Arabia’s digital-government transformation program.
- The agreement covers research, talent exchange and sandbox testing, giving government technology teams a framework for moving emerging AI concepts toward practical validation.
- DGA’s sandbox could help evaluate AI applications before production, addressing technical, security and governance challenges associated with deploying emerging technologies in government.
- Researcher exchanges and internships address the AI talent gap, strengthening connections between researchers, government teams and technology organizations across Hong Kong and Saudi Arabia.
- The partnership reflects Saudi Arabia’s broader AI ambitions, positioning digital government as an important testing ground for emerging artificial intelligence applications.
Power Tomorrow’s Intelligence — Build It with TechEdgeAI
