DUBAI, 1st June, 2026 (WAM) — Positron AI Ltd, a US-based developer of next-generation specialised AI inference infrastructure, has established its first presence outside the United States in the Dubai International Financial Centre (DIFC), the leading global financial centre in the Middle East, Africa and South Asia region.
The DIFC licence marks a key milestone in Positron’s regional expansion. The company has raised more than US$300 million, including a US$230 million Series B round, deployed inference technology with a US hyperscaler, and achieved unicorn status.
The move comes as AI inference becomes the dominant driver of global compute demand. AI inference refers to the deployment, operation and scaling of large language models reliably, economically and within real-world constraints.
Husni Khuffash, Managing Director MENA, Positron AI, said, “We are pleased to move to DIFC, a dynamic hub that is reflective of our commitment to innovation and pushing the boundaries of AI infrastructure. Positron AI’s focus on power-efficient, deployable inference solutions will help strengthen the wider ecosystem as demand accelerates for scalable AI capabilities. Establishing in DIFC positions Positron to engage closely with industry, regulators and partners, while contributing to Dubai’s ambition to build globally competitive digital and AI capabilities.”
Mohammad Alblooshi, Chief Executive Officer of the DIFC Innovation Hub, said, “We are pleased to welcome Positron AI to the Dubai AI Campus’ vast ecosystem. Positron AI brings a compelling approach to AI inference infrastructure that supports the next phase of growth across Dubai’s technology and innovation landscape. The company’s foundational ideology aligns with DIFC’s evolution into an AI-Native jurisdiction and destination. Their expansion positions the Centre as a launchpad enabling scale and a gateway that empowers future industries, setting a global benchmark for AI governance and responsible innovation.”
As AI demand shifts structurally from model training to scalable model inference, Positron enables lower cost per token, higher memory density and improved energy efficiency, positioning it as a next-generation alternative to traditional AI hardware providers. Its proprietary infrastructure is designed to address key industry challenges, including high power consumption, memory constraints and the rising costs associated with GPU-based systems.
The company’s first-generation server, Atlas, was designed to deliver LLM inference for small and medium-sized models, with up to 500 billion active parameters, at a performance level similar to DGX-H100 systems but with lower power consumption and cost.
Positron’s next-generation product, Titan, is scheduled for delivery in the first quarter of 2027 and will expand the company’s capabilities to run frontier models with greater performance and power efficiency than NVIDIA’s Rubin and Blackwell GPU systems.
Source: Emirates News Agency