Core42 Enhances AI Deployment for UAE Government Services with Secure Infrastructure

Core42 Enhances AI Deployment for UAE Government Services with Secure Infrastructure

Core42 is advancing the deployment of artificial intelligence (AI) within UAE government services by establishing a secure and scalable infrastructure. This initiative is crucial as the UAE aims to transition 50% of its government services and operations to Agentic AI, which necessitates a robust framework capable of delivering intelligence securely and economically at production scale. Raghu Chakravarthi, Chief Product and Technology Officer at Core42, emphasized the importance of sovereignty, compute flexibility, workload orchestration, observability, and financial governance in this transformative process.

Building a Sovereign AI Infrastructure

In an interview with Tahawultech.com, Chakravarthi outlined how integrated compute, cloud, data, and governance capabilities can facilitate the transition from isolated AI pilots to comprehensive government-wide deployment. He noted that the unique approach in the Gulf region is to pursue sovereignty and scalability simultaneously, ensuring that the infrastructure is locally governed and globally competitive from the outset.

Chakravarthi stated, “For government, the priority is therefore to embed scalability, sovereignty, observability, and governance into the inference layer so these capabilities can expand alongside Agentic AI rather than being added later.” This foundational strategy is essential as autonomous agents generate multiple inference events to complete complex tasks, which can significantly impact operational costs.

Optimizing AI Workloads and Cost Management

To effectively manage the diverse demands of AI workloads, Core42 is developing a workload classifier that will automatically route requests to the most suitable models and accelerators. This approach allows for better performance and cost control while ensuring sensitive government data remains protected. The deployment architecture will support flexibility across various processing environments, including real-time and batch processing, cloud, and on-premises solutions.

Chakravarthi highlighted the necessity of embedding data sovereignty directly into the inference layer, incorporating in-country data residency, encryption, and access controls as integral components of the deployment path. This proactive approach is supported by over 170 security policies and SOC 2 Type II certification, ensuring that government entities can maintain oversight and governance over their AI operations.

Future Implications for Government AI Services

As the UAE government prepares for broader AI adoption, the focus will be on managing the economics of the complete agentic workflow rather than merely counting users or prompts. Effective cost controls, including spend caps and budget alerts, must be established before scaling AI consumption. This strategic planning will help prevent unexpected expenses as AI systems become more autonomous and complex.

Ultimately, the integration of workload-aware orchestration across different models and accelerators will enhance the performance, sovereignty, and economic viability of government AI services. By treating infrastructure as a dynamic asset rather than a fixed choice, Core42 aims to ensure that each task is matched with the most appropriate resources, thereby improving efficiency and responsiveness in citizen-facing applications.