Table of Contents
Emma Cloney, senior vice president of international sales and strategy at Core42 told RCRTech that the firm addresses “data, operational, and technological sovereignty through in-country control of data, infrastructure, and models.”
In sum – what to know
Production, not infrastructure alone – Core42 says the biggest hurdle for enterprise AI is deploying secure, governed AI systems rather than simply accessing GPU capacity.
Sovereign AI partnership expands – Core42 and e& UAE have recently launched a sovereign AI compute platform combining in-country GPU infrastructure, connectivity and implementation services.
Inference workloads expected first – Financial services, healthcare, and other regulated sectors are expected to lead adoption before more advanced agentic AI applications mature.
Core42, a subsidiary of Abu Dhabi-based G42, believes the biggest challenge facing enterprise AI is no longer securing compute infrastructure, but moving AI projects into secure production deployments, as organizations face data governance, compliance, and operational integration challenges.
The company outlined that position while discussing its recently announced partnership with e& UAE, which aims to provide enterprises and government organizations with sovereign AI infrastructure hosted within the United Arab Emirates (UAE). The collaboration combines Core42’s Sovereign AI Cloud with e& UAE’s national digital infrastructure, secure connectivity and professional services to provide organizations with access to in-country GPU infrastructure for AI development and deployment.
According to Emma Cloney, senior vice president of international sales and strategy at Core42, the company’s approach to sovereign AI extends beyond the physical location of infrastructure.
“Core42’s differentiation lies in the depth and operability of sovereignty – a spectrum defined by each customer’s requirements – rather than ensuring the entire infrastructure stack is located in one place,” Cloney told RCRTech.
She added that Core42 addresses “data, operational, and technological sovereignty through in-country control of data, infrastructure, and models,” while allowing customers to choose the hardware best suited to their workloads and maintain visibility over where processing takes place.
The company also highlighted the importance of security and compliance for regulated industries. “Data residency in the UAE is complemented with strong encryption, SOC2 Type II compliance across 170 policies and zero data logging, providing regulated industries with an easy route to deploy and scale generative AI without compromising on data protection or compliance.”
The recently launched Sovereign AI Compute platform combines Core42’s sovereign AI cloud infrastructure with e& UAE’s connectivity and customer services under a single commercial and operational framework. According to the companies, the platform is designed to allow organizations to build, train, fine-tune, infer, and deploy AI while keeping sensitive workloads and data within the UAE.
Cloney said organizations frequently discover that deploying AI into production involves much more than securing compute capacity. “The hardest part for customers is rarely securing GPUs, although the capital expenditure required to run infrastructure can be prohibitive,” the executive said.
Instead, she said organizations face challenges around connecting governed enterprise data, selecting and optimizing AI models, integrating them into operational workflows, and maintaining security and compliance.
To address those issues, Core42 said its platform combines sovereign compute with high-bandwidth connectivity while providing access to more than 50 AI models together with implementation tools.
“Its full-stack, multi-silicon sovereign AI Cloud provides access to over 50 leading models with the tools to implement guardrails, regulate access, and scale inference, whilst removing the capital expenditure and long lead times associated with on-premises deployment,” said Cloney.
Looking ahead, Core42 expects adoption to begin with inference workloads in regulated sectors before progressing toward more advanced AI agents.
“Over the next 12-24 months, adoption is likely to begin with high-value, data-sensitive inference workloads rather than fully autonomous agents,” the executive added.
The company said early deployments are expected to focus on financial services—including automation, real-time data analysis, anti-fraud and AML reporting—and healthcare applications such as diagnostics and medical imaging.
For agentic AI, Cloney said organizations are likely to begin with human-supervised systems before expanding to more autonomous workloads.
“As deployments mature and organizations have the confidence to wield agents in a safe, secure and compliant manner, these applications will evolve to take on more complex, multi-step workloads.”
