October 15, 2025 · 24-Hour AI Briefing: Microsoft, Alibaba Cloud, and Oracle Accelerate the Global AI Infrastructure Race
The global AI race is shifting from model innovation to compute power.
This week, Microsoft, Alibaba Cloud, and Oracle each made major moves in data center expansion and chip deployment — signaling a new phase in the battle for AI infrastructure dominance.
1. Microsoft Rents Data Center Capacity in Portugal to Deploy Over 12,000 NVIDIA GPUs
Microsoft plans to lease data center capacity in Portugal from the UK-based hyperscale compute provider Nscale.
The facility will deploy 12,600 NVIDIA Ultra GPUs and is expected to go online in early 2026.
Nscale recently raised $1.1 billion in Series B funding and has partnered with Microsoft, NVIDIA, and OpenAI.
Commentary:
By renting rather than building, Microsoft is expanding its global GPU footprint at unprecedented speed — strengthening Azure’s backbone for Copilot, Microsoft 365 AI, and Azure OpenAI.
This marks a clear boost for Europe’s AI ecosystem, potentially stimulating regional startups and academic research.
However, Microsoft must still navigate EU regulations on data sovereignty and cross-border compute infrastructure.
2. Alibaba Cloud Opens Its Second Data Center in Dubai, Targeting the Middle East AI Market
Alibaba Cloud has launched its second data center in Dubai to meet the region’s growing demand for cloud and AI services.
With this expansion, Alibaba Cloud’s global network now covers 29 regions and 92 availability zones, spanning Thailand, South Korea, Malaysia, and Mexico.
Commentary:
Localized deployment allows Alibaba Cloud to deliver low-latency and highly compliant AI solutions for governments and enterprises.
The Dubai center focuses on AI training and inference workloads, powering industries such as finance, energy, and public services.
It is Alibaba Cloud’s largest infrastructure upgrade in the Middle East since 2017.
Yet, navigating the region’s regulatory complexity and competing with AWS and Azure will remain ongoing challenges.
3. Oracle to Deploy 50,000 AMD MI450 Chips, Joining the Push for AI Chip Diversification
Oracle plans to deploy 50,000 AMD MI450 AI chips in its data centers starting in Q3 2026.
These systems will integrate AMD’s CPUs and networking modules for large-scale AI workloads.
Commentary:
In a field long dominated by NVIDIA, Oracle’s choice of AMD marks a deliberate shift toward a more diverse AI supply chain.
This move challenges NVIDIA’s market dominance while balancing cost, performance, and security priorities.
For AMD, coming off its strategic deal with OpenAI, Oracle’s order could represent billions in revenue — accelerating AMD’s evolution from a CPU maker to a full-stack AI infrastructure provider.
Conclusion
The simultaneous expansion of Microsoft, Alibaba Cloud, and Oracle reveals a new phase of global competition.
The future of AI will not only depend on who builds the smartest model, but on who controls the compute, energy, and ecosystem balance behind it.
From the U.S. to the Middle East, compute power is rapidly becoming the “new energy gold” of the digital age.
For more AI insights and tech updates, visit:
👉 https://iaiseek.com/en
Recent highlights:
October 14, 2025 · 24-Hour AI Briefing: OpenAI Designs Its Own Chips, Samsung Surges on AI Boom, and ABB-NVIDIA Redefine Data Center Power