AI Hardware

Twitter

Facebook

Copy Link

Key Facts

Key PlayersNVIDIA, AMD, Google, Cerebras

Top News

KLA Corp projects strong quarter as AI infrastructure spending fuels semiconductor demand

KLA Corp reported fourth-quarter revenue of $3.66 billion and forecast first-quarter revenue up to $4.2 billion, surpassing Wall Street estimates. The semiconductor equipment manufacturer expects accelerating momentum through 2027, driven by AI infrastructure buildout and rising demand for process control systems. Despite the upbeat quarter forecast, shares fell 8% in after-hours trading.

Business2 sources

KLA Corp projects strong quarter as AI infrastructure spending fuels semiconductor demand

Business2 sources

US Robot Ban Targets Chinese Humanoid Robots, Escalating US-China Trade Tensions

The Federal Communications Commission banned foreign-made advanced robots including humanoid robots and quadruped robots, citing national security risks and data collection concerns. The move primarily targets Chinese robotics companies like Unitree, which dominate 90% of the global market, while China threatens retaliation against what it calls discriminatory AI protectionism.

Policy34 sources

US Robot Ban Targets Chinese Humanoid Robots, Escalating US-China Trade Tensions

Policy34 sources

Honor Robot Phone launches August 12 with dual 200MP cameras and cinema-grade AI imaging

Honor confirms its Robot Phone will launch August 12—the same day as Google's Pixel 11—featuring dual 200MP cameras mounted on a 4DoF gimbal system and professional ARRI cinema workflows. The device introduces Honor's AiMAGE architecture and custom Yuguang H1 imaging chip, marking the company's transition into an AI terminal ecosystem provider. Pre-orders have already surpassed 200,000 units in China.

Technology2 sources

Honor Robot Phone launches August 12 with dual 200MP cameras and cinema-grade AI imaging

Technology2 sources

Seagate's Hard Drives See Massive Demand as AI Storage Boom Extends Capacity Commitments to 2028

Seagate reports cloud providers now account for 90% of its exabyte shipments as AI infrastructure boom drives unprecedented demand for high-capacity hard drives. The company has already allocated most nearline drive capacity through calendar 2028, with customers now planning storage requirements into 2029.

Business7 sources

Seagate's Hard Drives See Massive Demand as AI Storage Boom Extends Capacity Commitments to 2028

Business7 sources

Mystery reviewer tests Nvidia RTX Spark prototype in Microsoft Surface Laptop Ultra

A TechPowerUp forum user spent a month with an unreleased Microsoft Surface Laptop Ultra powered by Nvidia's RTX Spark N1X chip. The pre-release AI hardware prototype showed potential but suffered from immature drivers, GPU stuttering, and power management issues. Performance fell short of the promised RTX 4070 levels, landing closer to RTX 3060 Ti territory.

Technology3 sources

Mystery reviewer tests Nvidia RTX Spark prototype in Microsoft Surface Laptop Ultra

Technology3 sources

Taiwan detains Nvidia employee in expanding AI chip smuggling probe targeting Super Micro servers

Taiwan prosecutors detained an Nvidia employee suspected of falsifying documents to smuggle AI servers to China, marking the first known legal action against an Nvidia staffer in such a case. The Keelung District Prosecutors' Office searched the employee's home and Taipei office desk, citing flight risk and evidence tampering concerns. Seven people now sit in custody as the investigation expands beyond Super Micro into Nvidia itself.

Policy10 sources

Taiwan detains Nvidia employee in expanding AI chip smuggling probe targeting Super Micro servers

Policy10 sources

Zhongji Innolight drops 9% in Hong Kong debut despite raising $6.8 billion in city's largest IPO

Chinese AI data center supplier Zhongji Innolight fell as much as 9.4% in its Hong Kong debut after raising $6.8 billion in the city's largest IPO since Alibaba's 2019 listing. The optical transceivers maker, which supplies both US tech giants and Chinese firms, was swept up in a global tech sell-off despite strong demand from hyperscalers building AI infrastructure.

Business7 sources

Zhongji Innolight drops 9% in Hong Kong debut despite raising $6.8 billion in city's largest IPO

Business7 sources

Cadence Design Systems raises annual forecasts as AI chip design demand surges by 24%

Cadence Design Systems increased its 2026 revenue and profit projections, driven by accelerating demand for AI-powered chip and system design software. The company now expects revenue between $6.26 billion and $6.34 billion, up from prior guidance of $6.13 billion to $6.23 billion. Second-quarter revenue jumped 24% year-over-year to $1.584 billion, with a record backlog of $8.1 billion signaling sustained momentum in semiconductor design tools.

Business4 sources

Cadence Design Systems raises annual forecasts as AI chip design demand surges by 24%

Business4 sources

AI vendors shift to consumption pricing as AI PCs emerge to control spiralling cloud costs

Major AI vendors are abandoning per-seat subscriptions for token consumption pricing, making cloud AI significantly more expensive for enterprises. AI PCs with neural processing units now offer a cost-effective alternative by running local models for routine tasks, providing zero marginal cost per query after hardware purchase. The shift creates a new hybrid model where enterprises balance cloud infrastructure for complex tasks against local compute for high-volume, low-complexity work.

Business2 sources

AI vendors shift to consumption pricing as AI PCs emerge to control spiralling cloud costs

Business2 sources

CXMT shares surge 466% in blockbuster $8.6 billion IPO as China's memory chipmaker rides AI boom

ChangXin Memory Technologies became mainland China's most valuable company after its shares rocketed over 466% on debut, raising $8.6 billion in the country's biggest IPO since 2010. The DRAM memory chipmaker's meteoric rise reflects surging AI-driven demand, but faces challenges from U.S. export restrictions and geopolitical tensions.

Business10 sources

CXMT shares surge 466% in blockbuster $8.6 billion IPO as China's memory chipmaker rides AI boom

Business10 sources

Old Nvidia GPUs with massive VRAM are dominating local AI inference over newer, faster cards

Computing enthusiasts are discovering that older Nvidia GPUs like the Tesla V100 and RTX 3090 outperform newer cards for local AI inference, thanks to their abundant VRAM. One modder added a $266 Tesla V100 to their gaming PC, achieving 32GB of total VRAM to run a 27 billion parameter model at 32 tokens per second. The trend highlights how memory capacity matters more than raw compute power for running LLMs at home.

Technology4 sources

Old Nvidia GPUs with massive VRAM are dominating local AI inference over newer, faster cards

Technology4 sources

Samsung and Broadcom forge $200 billion AI chip partnership to challenge TSMC's dominance

Samsung Electronics signed a $200 billion memorandum of understanding with Broadcom to supply AI chips, memory, and foundry services through 2030. The Samsung Broadcom partnership will leverage sub-2-nanometer manufacturing and advanced packaging to produce custom AI accelerators, strengthening Samsung's position against industry leader TSMC.

Business7 sources

Samsung and Broadcom forge $200 billion AI chip partnership to challenge TSMC's dominance

Business7 sources

NAVER, NVIDIA and Brookfield forge $10 billion partnership to build Korea's sovereign AI future

NVIDIA is investing $1 billion to acquire a 4.5% stake in NAVER Corp, while Brookfield commits up to $9 billion to expand South Korea's AI infrastructure. The partnership will scale NAVER's GAK Sejong data center from 55 megawatts to 200 megawatts by 2028, with ambitions to reach gigawatt-scale capacity. This marks a strategic push to build sovereign AI infrastructure that can fuel Korea's AI ecosystem and compete globally.

Technology10 sources

NAVER, NVIDIA and Brookfield forge $10 billion partnership to build Korea's sovereign AI future

Technology10 sources

OpenAI's first hardware device sold out in 12 hours, now reselling for up to $1,850 on eBay

OpenAI launched its debut hardware product, the Codex Micro AI keypad, at $230 and it vanished from shelves in just 12 hours. The programmable macro pad for coders now commands premium prices on eBay, with some listings reaching $1,850. Built in collaboration with Work Louder, the device offers a physical interface for Codex but has sparked debate about whether it's worth the investment.

Technology3 sources

OpenAI's first hardware device sold out in 12 hours, now reselling for up to $1,850 on eBay

Technology3 sources

Nvidia and SK Group forge $500 billion AI partnership spanning data centers and memory chips

Nvidia has locked in a massive $500 billion partnership with South Korea's SK Group to secure high-bandwidth memory supply and build large-scale AI data centers. SK Telecom will construct a 2-gigawatt facility using Nvidia's Vera Rubin chips and SK Hynix's HBM4 memory, with operations starting in 2027. Separately, Nvidia is investing $1 billion in Naver to expand AI infrastructure in South Korea.

Technology8 sources

Nvidia and SK Group forge $500 billion AI partnership spanning data centers and memory chips

Technology8 sources

Prev 1…789…99 Next