
Oracle Announces $50 Billion AI Infrastructure Build-Out
Oracle is raising up to $50 billion to fund an AI infrastructure build-out designed to power next-generation cloud and enterprise AI workloads.

Oracle is raising up to $50 billion to fund an AI infrastructure build-out designed to power next-generation cloud and enterprise AI workloads.

Microsoft introduced Maia 200, a 3nm inference accelerator with 216GB HBM3e and large on-chip SRAM, aimed at lowering AI serving costs and power use.

NVIDIA unveiled the Vera Rubin platform at CES 2026, combining the Rubin GPU and Vera CPU to reduce inference costs and support robotics and autonomous systems.

The US approved Nvidia H200 exports to China, but Beijing is preparing its own limits, reshaping what access really means.

Microsoft is committing more than $10B to expand AI infrastructure in Portugal, adding 12,600 Nvidia GPUs to the Sines hub and strengthening European compute capacity.

Nvidia’s $5tn milestone week includes a landmark APEC-announced plan to deliver Blackwell AI chips to South Korea’s sovereign clouds and AI factories, with telecom, quantum, and startup programs expanding the ecosystem.

Apple unveils the M5 chip with 4x AI GPU gains, 45% graphics uplift, 16-core Neural Engine, and 153GB/s memory bandwidth in MacBook Pro, iPad Pro, and Vision Pro.

OpenAI and Broadcom will jointly design custom AI chips and rack systems for a 10 GW rollout by 2029. The deal complements Nvidia, AMD, and Oracle partnerships, solidifying OpenAI’s vertical AI infrastructure strategy.

OpenAI secured a multi-year partnership with AMD for 6GW of AI compute, starting in late 2026. The deal links OpenAI to nearly 10% of AMD equity and sparked a record stock surge.
Automated page speed optimizations for fast site performance