llms.blog
AI News

Nvidia and Amazon are spending billions to power the AI buildout

Nvidia and Amazon are spending billions to power the AI buildout The AI boom has outgrown the grid. Nvidia and Amazon are now putting billions of dollars into power generation and infrastructure, with consequences for both the supply chain and the climate. Nvidia is investing up to $3 billion in Lancium, the power-infrastructure developer behind the OpenAI-Oracle Stargate site in Texas, according to The Information. A $2 billion stake would give Nvidia roughly 20 percent of a company valued at

axc
axc
2 min
Nvidia and Amazon are spending billions to power the AI buildout

Nvidia and Amazon are spending billions to power the AI buildout

The AI boom has outgrown the grid. Nvidia and Amazon are now putting billions of dollars into power generation and infrastructure, with consequences for both the supply chain and the climate.

Nvidia is investing up to $3 billion in Lancium, the power-infrastructure developer behind the OpenAI-Oracle Stargate site in Texas, according to The Information. A $2 billion stake would give Nvidia roughly 20 percent of a company valued at about $10 billion. Lancium already has four gigawatts of power under contract in Texas and is developing sites for up to 15 more gigawatts.

Amazon is going further downstream. It is backing a gas-fired plant in Pecos County, Texas, designed to feed up to 7.65 gigawatts to an Amazon data center through 35 gas turbines. The New York Times reports the facility could emit up to 33 million tons of CO2 a year, which would make it the dirtiest power plant in the United States. Amazon spokesperson Margaret Callahan said the company's climate goals still stand but acknowledged that AI data centers could make them harder to hit.

The scale reflects how quickly compute demand is outpacing the grid. Climate scientist Zeke Hausfather estimates that a heavy user of agentic AI burns roughly the same energy per year as a clothes dryer, and that figure keeps climbing as agents take on more tasks.

The investments show where the money is flowing in the AI race: not just chips, but the electricity to run them. Nvidia's move into power infrastructure also tightens its grip on the entire stack, from silicon to the watt.

Sources

AI NewsPolicy
ShareX (Twitter)LinkedIn

Written by

axc

More to read

  • Activation Checkpointing in Large Language Models: How Selective Recomputation Eliminates Memory Bottlenecks

    Large language model pre-training and fine-tuning are fundamentally constrained by GPU memory (VRAM). While distributed techniques such as Fully Sharded Data Parallel (FSDP), ZeRO, and Tensor Parallelism successfully shard model parameters, optimizer states, and gradients across hundreds or thousands of GPUs, activation memory presents a distinct scaling bottleneck. During the forward pass of a transformer model, intermediate tensor outputs must be preserved in GPU memory so that backpropagatio

    1 min

  • Anthropic Demonstrates Autonomous De Novo Protein Design and Chemical Analysis with Claude

    Anthropic Demonstrates Autonomous De Novo Protein Design and Chemical Analysis with Claude Anthropic has published experimental results demonstrating Claude's ability to autonomously design de novo protein binders with physical wet-lab validation and automate complex analytical chemistry workflows. The findings show frontier LLMs acting as autonomous agents across computational biology and molecular characterization pipelines. In the primary experiment, Anthropic evaluated Claude Mythos Previe

    1 min

  • Cerebras Unveils CS-4 Rack-Scale System Powered by Three WSE-3 Turbo Chips and Nexus Architecture

    Cerebras Unveils CS-4 Rack-Scale System Powered by Three WSE-3 Turbo Chips and Nexus Architecture Cerebras Systems has announced the CS-4, a rack-scale AI accelerator system designed around three of its next-generation Wafer Scale Engine 3 Turbo (WSE-3 Turbo) chips and a modular hardware architecture dubbed Nexus. Cerebras confirmed that initial customer shipments for the CS-4 are scheduled to begin in the current quarter. The new system marks a structural shift from Cerebras's single-wafer CS

    1 min