AI Hardware

Twitter

Facebook

Copy Link

Key Facts

Key PlayersNVIDIA, AMD, Google, Cerebras

Top News

NVIDIA Open-Sources cuFile API and Launches Storage-Next Initiative With 40+ Industry Partners

NVIDIA announced it is open-sourcing its cuFile API at the Future of Memory and Storage conference, enabling direct GPU-to-storage data access in microseconds. The company also launched Storage-Next, bringing together over 40 storage vendors including DDN, Kioxia, and Micron to standardize GPU-driven storage solutions for AI workloads.

Technology4 sources

NVIDIA Open-Sources cuFile API and Launches Storage-Next Initiative With 40+ Industry Partners

Technology4 sources

Sharon AI Secures $373 Million Cloud Deal as Pipeline Hits $8.8 Billion Despite Q2 Revenue Miss

Sharon AI announced a $373 million cloud computing service agreement with a global AI platform, bringing its contracted AI factory capacity to 120MW. The deal comes as the Australian AI infrastructure company reported Q2 revenue of $1.93 million against analyst expectations of $7.54 million, while simultaneously revealing an $8.8 billion total contract value pipeline.

Technology4 sources

Sharon AI Secures $373 Million Cloud Deal as Pipeline Hits $8.8 Billion Despite Q2 Revenue Miss

Technology4 sources

AI Power Demand Crisis Forces Data Centers to Buy Ship Engines for Off-Grid Solutions

A South Korean shipbuilder sold $673.8 million worth of engines to power US data centers off the grid, marking AI's escalating power crunch. HD Hyundai's 1,000 MW engine order highlights how grid interconnection delays are forcing tech companies to build their own power generation, even as volatile power demand damages critical infrastructure.

Technology8 sources

AI Power Demand Crisis Forces Data Centers to Buy Ship Engines for Off-Grid Solutions

Technology8 sources

US Ban on Chinese AI Hardware Could Disrupt Data Centers and Raise Costs for Amazon, Microsoft

The Trump administration is drafting a ban on Chinese data center components, targeting optical transceivers that power AI infrastructure. Zhongji Innolight and Eoptolink shares plunged 8-16% as the Federal Communications Commission moves to block imports over national security concerns, threatening to disrupt AI supply chains and increase costs for major hyperscalers.

Policy14 sources

US Ban on Chinese AI Hardware Could Disrupt Data Centers and Raise Costs for Amazon, Microsoft

Policy14 sources

OpenAI Calls Apple's Trade Secrets Lawsuit 'Careless and Oddly Personal' as Legal Battle Escalates

OpenAI has fired back at Apple's trade secrets lawsuit, calling it 'careless, aggressive and oddly personal' while filing a motion to dismiss. The AI company argues Apple's own security lapses allowed former employees access to systems, and that more than 400 ex-Apple workers now at OpenAI were simply attracted by innovative AI work.

Business35 sources

OpenAI Calls Apple's Trade Secrets Lawsuit 'Careless and Oddly Personal' as Legal Battle Escalates

Business35 sources

Google Builds $200bn AI Chips Infrastructure With Broadcom, Apollo & Blackstone to Power Anthropic

Google has assembled one of the largest infrastructure financing programmes in history—worth roughly $200 billion—to sell Tensor Processing Units to Anthropic. The effort brings together Broadcom, Apollo, Blackstone, and Morgan Stanley in a web of transactions spanning chip manufacturing to data centre development, potentially unlocking $252 billion in TPU revenue by 2028.

Business2 sources

Google Builds $200bn AI Chips Infrastructure With Broadcom, Apollo & Blackstone to Power Anthropic

Business2 sources

Snap CEO Evan Spiegel: AI Glasses Mass Adoption Years Away Despite Strong Q2 Revenue Growth

Snap beat Q2 revenue expectations with $1.6 billion, up 19% year-over-year, as CEO Evan Spiegel positioned the company's $2,195 augmented reality glasses as its biggest bet on the future. Despite unveiling SPECS in September, mass market adoption won't arrive until the end of the decade as Snap pursues an AI-powered computing platform designed to reduce screen time.

Business2 sources

Snap CEO Evan Spiegel: AI Glasses Mass Adoption Years Away Despite Strong Q2 Revenue Growth

Business2 sources

Sandisk and SK hynix Release First HBF Spec: 512GB Capacity, 3 TB/s Bandwidth for AI Inference

Sandisk and SK hynix released the first High Bandwidth Flash technical specification through the Open Compute Project, defining a new memory tier for AI inference systems. The specification establishes capacities up to 512GB with bandwidth ranging from 0.4 TB/s to 3 TB/s, using UCIe interconnect to bridge the gap between HBM performance and NAND capacity.

Technology12 sources

Sandisk and SK hynix Release First HBF Spec: 512GB Capacity, 3 TB/s Bandwidth for AI Inference

Technology12 sources

ON Semiconductor Expects AI Data Center Revenue to Double in 2026 as Q2 Earnings Beat Estimates

ON Semiconductor reported second-quarter revenue of $1.6 billion, surpassing analyst expectations as demand for power management chips in AI data centers accelerated. CEO Hassane El-Khoury revealed the company prioritized AI data center customers over automotive clients, shifting manufacturing capacity to meet surging demand for intelligent power solutions.

Business5 sources

ON Semiconductor Expects AI Data Center Revenue to Double in 2026 as Q2 Earnings Beat Estimates

Business5 sources

Majestic Labs unveils Prometheus AI server with 128TB memory to challenge Nvidia's dominance

Tel Aviv startup Majestic Labs has unveiled Prometheus, a GPU-free server that uses up to 128TB of LPDDR6 memory and Arm cores to tackle AI's memory wall. The company claims one rack matches 25 Nvidia Vera Rubin racks at lower power, but hardware hasn't shipped or been independently tested yet.

Technology2 sources

Majestic Labs unveils Prometheus AI server with 128TB memory to challenge Nvidia's dominance

Technology2 sources

James Dacombe's Olix AI Chip Startup Triples Valuation to $3.3bn in Six Months

James Dacombe's AI chip startup Olix raised $312mn from Fundomo, Arm, and Reed Hastings, tripling its valuation to $3.3bn in six months. The London-based firm bypasses supply chain bottlenecks with photonic interconnect technology, positioning itself as a specialized alternative to Nvidia in the booming AI infrastructure market.

Startups5 sources

James Dacombe's Olix AI Chip Startup Triples Valuation to $3.3bn in Six Months

Startups5 sources

Nothing Pivots to AI Devices as Smartwatch, Earbuds & Speakers Leak Ahead of Reported August Launch

Nothing is reportedly making a strategic shift from smartphones to AI devices, with leaked images showing a smartwatch, CMF Ear Clip Buds, headphones and speakers. The move comes as co-founder Carl Pei backs the pivot to AI-powered accessories and wearables, potentially launching in August or September.

Technology2 sources

Nothing Pivots to AI Devices as Smartwatch, Earbuds & Speakers Leak Ahead of Reported August Launch

Technology2 sources

Memory Chip Shortage to Intensify Through 2027 as AI Demand Pushes Prices Up 5-6x

Samsung warns the global shortage of RAM chips will deepen in 2027 and persist until 2028 as AI-driven demand overwhelms supply. Memory chip prices have surged 5-6 times, forcing Apple and Samsung to raise device prices while manufacturers like SK Hynix invest $38 billion in new facilities that won't open until 2028.

Business8 sources

Memory Chip Shortage to Intensify Through 2027 as AI Demand Pushes Prices Up 5-6x

Business8 sources

NXP Semiconductors eyes $3.3bn Ambarella acquisition to dominate AI vision chips for autonomous vehicles

NXP Semiconductors is negotiating to acquire Ambarella for approximately $3.3 billion to strengthen its position in AI-enabled vision processing for self-driving cars. The deal would integrate Ambarella's edge AI chips into NXP's automotive portfolio, which already accounts for roughly half of its projected $15.1 billion revenue in 2026.

Business7 sources

NXP Semiconductors eyes $3.3bn Ambarella acquisition to dominate AI vision chips for autonomous vehicles

Business7 sources

Qualcomm Humanoid Robot Collapses Face-First During Live Demo at Computex 2026

A Qualcomm-powered humanoid robot collapsed spectacularly on stage at Computex 2026 in Taiwan, falling face-first during a live demonstration. The 4NE1 Gen 3.5 robot from Neura Robotics, running on Qualcomm's Dragonwing IQ10 platform with 700 TOPS of AI processing power, had to be covered and carried away after the malfunction.

Technology2 sources

Qualcomm Humanoid Robot Collapses Face-First During Live Demo at Computex 2026

Technology2 sources

Prev 1…171819…111 Next