Chips, GPUs, data centers, cloud compute, and edge AI hardware
41 stories in the last 7 days
Accelerate agentic AI with NVIDIA Nemotron models on Amazon EKS and SageMaker HyperPod EKS
NVIDIA Nemotron 3.5 Super VL, a 120B‑parameter mixture‑of‑experts model, is now deployable on AWS EKS and SageMaker HyperPod EKS. The pos…
India’s Enterprises are Scaling AI. Is the Infrastructure Ready?
India’s enterprises are scaling AI, but legacy systems and fragmented data threaten smooth production rollouts. The article highlights in…
AI has come for Arizona’s water
TSMC’s new Arizona foundry consumes nearly 5 million gallons of water daily, fueling the state’s chip production boom. The plant’s massiv…
The Economics of Continuous Inference
Inference cost drives AI deployment economics. The article explains how real‑world usage turns a model’s performance into a compute budge…
How to Evaluate CPU Performance for Agentic AI
How to Evaluate CPU Performance for Agentic AI is a guide that benchmarks CPU efficiency for autonomous AI agents. It explains key metric…
Intel’s $2.3 Billion Bet on a Quantum Future is a Decade in the Making
Intel pours $2.3 billion into quantum computing, signaling a decade‑long push toward scalable qubit systems. The investment targets advan…
Akamai Launches AI Inference and Agent Sandbox Services
Akamai launches AI inference and agent sandbox services to provide a unified platform for building, running, and securing AI applications…
China's CXMT to start advanced AI memory production despite poor yields
CXMT plans to begin HBM3E production for AI processors despite low yields. Only 20‑30% of chips meet quality standards, yet the company p…
L&T’s Vyoma.AI Breaks Ground on 52 MW Data Centre in Chennai
L&T’s Vyoma.AI opens a 52‑MW data centre in Chennai that uses direct‑to‑chip liquid cooling to manage heat from high‑density workloads. T…
OpenAI Plans to Purchase Google TPUs: Report
OpenAI plans to buy Google TPUs to cut dependence on NVIDIA and scale training capacity. The move signals a shift toward alternative hard…
Elon Musk says more than 200 neoclouds face ‘guppy-level death’ - What it means for AI infrastructure push
Elon Musk warns that over 200 neoclouds could fail as debt financing dries up, challenging the AI infrastructure boom. He argues that wit…
Oracle Expands AI Compute Options With NVIDIA, Google TPUs and Cerebras
Oracle expands its AI compute portfolio, adding NVIDIA GPUs, Google TPUs, and Cerebras systems, while also integrating SambaNova inferenc…
OpenAI to purchase Google's AI chips amid rising demand for AI computing resources
OpenAI plans to buy Google’s AI chips to boost its computing power. The deal would give OpenAI access to Tensor Processing Units, a cheap…
The data center boom has an upside: innovation for the US grid
The AI-driven data center boom strains the U.S. power grid, prompting new grid modernization opportunities. Experts note that increased d…
Code-execution flaw in UEFI firmware affects most servers on planet Earth
A critical flaw in the EDK2 UEFI firmware lets attackers run code during early boot. The vulnerability, affecting most servers worldwide,…
Oracle is Expanding the Accelerator Options in OCI for Enterprises
Oracle expands accelerator options in OCI, adding GPUs and custom ASICs to boost enterprise workloads. Developers can now choose from NVI…
Announcing SOCI-indexed images for faster SageMaker AI jobs
SOCI indexing lets SageMaker skip full image downloads, starting containers in seconds. The new seekable OCI feature loads only needed la…
PrivacyLib: Privacy Controls at the Data Access Boundary
Meta introduces PrivacyLib, a shared library that enforces privacy rules across its data systems. The new component replaces duplicated l…
ANYbotics begins delivering ANYmal X legged robot for explosive atmospheres
ANYbotics delivers the ANYmal X legged robot, built to meet the highest safety standards for explosive atmospheres in oil, gas, and chemi…
Embotech and BMW move 1M vehicles with automated driving technology
Embotech deploys automated driving tech to move 3,500 vehicles daily across four BMW sites in the U.K. and Germany. The system operates 2…
Xpanner launches autonomous pile driver for solar farm contruction
Xpanner releases the X1 Piling R6, a robot that autonomously drives solar posts with a single operator. The machine improves placement ac…
How solid-state transformers could help power the AI boom
Solid‑state transformers could power the AI boom by boosting grid efficiency. These devices move electricity over long distances with low…
E2E Networks Plans 5.4 MW NVIDIA Blackwell Expansion in India
E2E Networks expands its AI data center in India with a 5.4 MW NVIDIA Blackwell deployment, boosting capacity by 59% to 14.6 MW. The roll…
Anthropic warns TSMC disruptions could hit AI chip supply, threaten business
Anthropic warns that disruptions at TSMC could choke its AI chip supply, limiting model training and deployment. The company cites geopol…
Anthropic is burning gas to feed the AI boom
Anthropic reveals its AI data centers depend on natural gas, exposing vulnerability to fuel price hikes and supply constraints. The S‑1 f…
OMRON to acquire OnLogic to address edge AI applications
OMRON acquires OnLogic to boost edge AI capabilities in industrial automation. The deal adds OnLogic’s industrial computing platform, ena…
SpaceX locks in $40B GPU financing deal with Apollo
SpaceX secures a $40 billion GPU financing deal with Apollo to boost its AI hardware supply. The partnership provides a steady stream of …
Meta Networking at the 2026 OCP Global Summit: Open Innovation for Scale-Up, Scale-Out, and Scale-Across AI Networks
Meta unveils liquid-cooled networking racks for its massive AI clusters. The IcePack and IceCube systems feature 102.4T switch blades usi…
Tata Communications To Pour Up To $450 Mn To Build AI Connectivity Infrastructure
Tata Communications invests $400‑450 million to build 10,000 km of AI‑ready connectivity across India. The rollout will upgrade network c…
Agility and GXO Logistics plan expanded deployment of Digit humanoid
Agility will deploy its Digit humanoid robot at two additional GXO Logistics sites, expanding its use in warehouse automation. The rollou…
Nvidia Simulates Its Way Into Google's Magic State Cultivation Party
Nvidia simulates its way into Google’s Magic State Cultivation Party. The company uses advanced GPU simulations to model quantum‑like sta…
India a preferred destination for data centres: Why world should invest in India, Vedanta chairman Anil Agarwal explains
India becomes a top choice for data centre investment, thanks to low construction costs, renewable energy prospects, and a huge consumer …
HCLTech Taps NVIDIA Simulation Tools to Automate Software-Defined Vehicle Testing
HCLTech adopts NVIDIA’s simulation tools to automate software‑defined vehicle testing. The partnership enables large‑scale, cloud‑based v…
Tata Communications Plans $450 Mn Investment to Build 10,000 Km AI Network
Tata Communications invests $450 million to build a 10,000 km AI network across India. The fiber‑optic backbone will link major data‑cent…
Wall Street's fastest traders are building data centers and backing chip startups to stay ahead in AI
High‑frequency traders are building data centers and investing in chip startups to secure AI compute. Jane Street and Hudson River Tradin…
Toyota is betting $6.4 billion a year on humanoid robots and factory automation
Toyota commits $6.4 billion annually to build a robotics empire, focusing on humanoid robots and factory automation. The investment aims …
Musk Says SpaceX Closed $40 Bn GPU Debt Deal, Warns Most Neoclouds Won’t Survive
SpaceX secures $40 billion debt to expand AI‑ready GPU infrastructure, leveraging Starlink and launch revenue. The deal positions the com…
Teradyne Robotics partners with Sereact to scale physical AI in retail and logistics
Teradyne Robotics teams with Sereact to deploy dual‑arm UR robots in retail and logistics sites. The partnership expands physical AI depl…
The great compute buildout: Can data centres power India’s AI leap?
India’s AI data‑center boom is backed by billions from global tech giants, aiming to turn the country into an AI creator. The buildout ta…
Will AI expose the rickety infrastructure of the internet?
AI tools probe the internet’s fragile backbone, revealing hidden bottlenecks and security gaps. Researchers use large‑language models to …
Rapidly scaling online storage to serve over 1 billion ChatGPT users (Part II)
OpenAI scaled its storage system Habitat to serve 1 billion users using Azure Cosmos DB and global caching. The system delivers 12x faste…