Susana Schwartz
Today at Yotta in Las Vegas, energized power capacity is the hot topic, with multi-year interconnection queues driving companies toward sites with high-voltage electricity already actively flowing or fully permitted to flow immediately.
Interesting is the fact that the void is being filled somewhat by companies that got their start in cryptomining, such as Onmine, whose CEO and co-founder Ahmed Elgamal said he bought up undervalued 10 MW, 20 MW and 30 MW sites, “while others were focusing on massive hyperscale campuses of 100+ MW” because he saw a future in which dedicated electrical substations and direct, ready-made connections to the electrical grid were going to be necessary to get GPUs up and running years faster than would be possible if building new sites from scratch. “We realize you’re losing revenue if your GPUs are sitting without space, so we find homes for the GPUs, offering immediate power,” he explained.
Read the story here, and check out other “top stories” and industry news below in “What you need to know.”
Susana Schwartz
Technology Editor
RCRTech
AI Infrastructure Top Stories
OpenAI – matching chips to workloads: Is OpenAI slowly but surely moving away from Nvidia? According to CFO Sarah Friar, the company is moving toward a multi-vendor compute and diversified supply chain strategy intended to mitigate risk.
AI Today: What You Need to Know
Goldman Sachs $1.2 Trillion CapEx Forecast: Goldman Sachs projects that Amazon, Alphabet, Microsoft, Oracle, and Meta will ramp up AI infrastructure spending from roughly $800 billion in 2026 to $1.2 trillion in 2027.
Microsoft 38 GW plan: Microsoft will expand its global DC capacity to more than 38 GWs by 2032 – 3x its current 12-GW footprint. The company added another GW of capacity in fiscal Q4 and is on track to double its overall capacity within 2 years.
LG transition from consumer to AI market: LG Electronics joined the NVIDIA Partner Network as a preferred provider, with its 2.6-MW Coolant Distribution Unit achieving “NVIDIA DSX Ready” status, with “chip-to-chiller” thermal infrastructure.
Firebird 70k Nvidia GPUs in Armenia: San Fran-based startup Firebird will build one of the largest computational clusters in Armenia, with its AI factory in its operational phase and on track to scale up to 300 MWs of capacity by end of 2027.
Nvidia RTX chips in China: China’s Ministry of Industry & Information Technology asked Alibaba and ByteDance to report purchasing plans for Nvidia’s RTX Pro 5500 chips, possibly to approve transactions that don’t violate U.S. DC export controls.
RCR Events
6G Forum, October 6th
The virtual 6G Forum will tackle the most pressing questions relating to AI-native network architectures and service enablement as the industry moves towards the 6G era. Register now
CCA Annual Convention, September 14-16th, New Orleans, Louisiana
Join industry stakeholders and innovative leaders in the communications service provider community this fall at CCA’s 2026 Annual Convention. Register now
RCR Roundtables AI Infrastructure, October 21st, Dallas, Texas
Join 50 senior data center, energy and AI leaders at the Ritz-Carlton Dallas on October 21 for invitation-only roundtables on powering and scaling AI. Request your invitation
Industry Resources
Report:Â Securing telecom infrastructure for the quantum era
Report: NTN in motion — evolving standards, expanding services
Market Pulse Report: Quantum safe networks  Report: Building AI-era network fabricsÂ
