Stories tagged by our AI curator.
Amazon has cancelled plans to build four new AI data centers in Northern Virginia following local opposition and backlash.
Google's data center expansion in Finland for AI operations has led to forest clearance, causing wolves to migrate into nearby villages.
Meta projects capital expenditure of $135-145 billion for 2026 to expand AI infrastructure and develop new products, with AI investments already generating incremental advertising revenue.
Enterprise AI success depends less on model selection and more on the engineering infrastructure—the serving harness—that delivers intelligence reliably and at scale in production environments.
Investors sold shares of Meta and other hyperscaler companies after Meta announced plans to issue new share class for capital raising to fund AI infrastructure buildout.
Google open-sourced TorchTPU and AWS launched TorchNeuron to help developers run PyTorch workloads on their chips as alternatives to Nvidia's GPU-dominated ecosystem.
Tech giants including Google, Amazon, and Meta are offering increased financial incentives and transparency pledges to win over communities opposing new data center construction.
Meta has committed to approximately $825 billion in future leases and contracts, primarily for data centers and computing infrastructure supporting its AI initiatives.
Chatterjee Group launches Cortex DC to develop 2 GW of AI-focused data centre capacity across India.
Nutanix announced infrastructure partnerships with NVIDIA, Supermicro and Netris to support production AI deployment, focusing on NVIDIA's Vera CPUs for agentic AI workloads.
Opinion piece argues that Anthropic's delayed IPO and conflicting market signals around AI safety concerns, rival CEO concerns, and the company's own prospectus warnings of existential risks suggest structural issues with the AI IPO wave.
Analysis reveals that AI inference operations generate substantial hidden data storage costs often overlooked in infrastructure planning, with backend data footprints significantly exceeding user-visible output sizes.
AI workloads are forcing data centers to adopt hybrid liquid and air cooling systems as GPU-accelerated infrastructure now routinely exceeds 40 kW per rack, far beyond air-only cooling's 20-40 kW threshold.
Dell reports that supply constraints in the AI infrastructure market now outweigh price considerations, shifting enterprise purchasing behavior.
Era4 addresses financial and data sovereignty concerns in Britain's AI infrastructure by building a U.K.-owned network, raising questions about control and resource distribution.
Three dormant US nuclear reactors are being restarted to provide dedicated power for artificial intelligence data center operations.
Nvidia CEO Jensen Huang unveiled the DSX Pod, a plug-in data center designed to address Europe's AI infrastructure gap, alongside plans for autonomous humanoid robotics.
Nvidia CEO Jensen Huang unveiled DSX POD, a containerized AI infrastructure system combining liquid-cooled Vera Rubin racks designed for rapid deployment and water-efficient operations at scale.
