AI Chips

Twitter

Facebook

Copy Link

Key Facts

Key ChipsH100, B200, TPU v5, Trainium

Key PlayersNVIDIA, AMD, Broadcom, Cerebras

Top News

Taiwan Semiconductor's Energy Crisis Could Disrupt Global AI Supply Chain

TSMC controls over 90% of advanced semiconductor manufacturing, but Taiwan's reliance on liquefied natural gas imports through the Strait of Hormuz creates a critical vulnerability. With LNG stockpiles lasting only days, the Middle East crisis threatens to halt chip production that powers Nvidia, AMD, and the entire AI industry.

Business2 sources

Taiwan Semiconductor's Energy Crisis Could Disrupt Global AI Supply Chain

Business2 sources

Intel bets billions on advanced packaging as Google and Amazon eye alternatives to TSMC bottleneck

Intel is in talks with Google and Amazon to provide advanced chip packaging services worth billions annually, positioning itself as an alternative to capacity-constrained TSMC. The chipmaker's CFO projects packaging revenue exceeding $1 billion, up from hundreds of millions, as AI demand creates a critical bottleneck in chip production. Intel's EMIB technology competes directly with TSMC's CoWoS as tech giants scramble for custom AI chip capacity.

Technology4 sources

Intel bets billions on advanced packaging as Google and Amazon eye alternatives to TSMC bottleneck

Technology4 sources

Intel's Neural Texture Compression slashes VRAM use by 18x as Nvidia shows 6.5GB reduced to 970MB

Intel unveiled Texture Set Neural Compression (TSNC), an AI-powered compression technology that can reduce game texture sizes by up to 18 times while maintaining image quality. Nvidia demonstrated similar capabilities with its Neural Texture Compression, cutting VRAM usage from 6.5GB to just 970MB in demo scenes. Both solutions address mounting storage and memory challenges as modern games demand increasingly detailed assets.

Technology6 sources

Intel's Neural Texture Compression slashes VRAM use by 18x as Nvidia shows 6.5GB reduced to 970MB

Technology6 sources

Apple Silicon Macs gain eGPU support for AI workloads as Apple approves TinyGPU driver

Apple has officially approved TinyGPU drivers from Tiny Corp, enabling AMD and Nvidia eGPU support on Apple Silicon Macs for the first time since the M1 launch. The drivers are designed exclusively for AI workloads, allowing users to run larger local AI models without bypassing system protections. Mac Mini owners running OpenClaw agents can now tap into external GPU power via Thunderbolt/USB4 connections.

Technology4 sources

Apple Silicon Macs gain eGPU support for AI workloads as Apple approves TinyGPU driver

Technology4 sources

DeepSeek V4 model to run on Huawei chips, testing China's AI self-sufficiency ambitions

DeepSeek's anticipated V4 model will run on Huawei chips instead of Nvidia hardware, marking a strategic shift in China's AI development. Chinese tech giants including Alibaba, ByteDance, and Tencent have ordered hundreds of thousands of Huawei chips ahead of the launch. The move signals China's push for AI self-sufficiency amid US export restrictions.

Technology5 sources

DeepSeek V4 model to run on Huawei chips, testing China's AI self-sufficiency ambitions

Technology5 sources

SpaceX files for IPO with $2 trillion valuation target, merging space exploration and AI ambitions

Elon Musk's SpaceX has confidentially filed for what could become the largest IPO in history, targeting a valuation exceeding $2 trillion. The company, now merged with xAI, plans to raise up to $75 billion to fund ambitious projects including orbital data centers, Starship development, and a lunar base.

Business3 sources

SpaceX files for IPO with $2 trillion valuation target, merging space exploration and AI ambitions

Business3 sources

Singapore charges third person in AI chip fraud case linked to Nvidia export violations

Singapore prosecutors charged Jenny Lim with fraud for allegedly misleading Dell Technologies about server purchases containing Nvidia chips. The case expands an ongoing investigation into the circumvention of U.S. export bans, with servers routed through Singapore to Malaysia and potentially China. Lim joins two co-conspirators previously charged in February 2025.

Policy5 sources

Singapore charges third person in AI chip fraud case linked to Nvidia export violations

Policy5 sources

IBM and Arm team up to bring modern AI workloads to enterprise mainframes via virtualization

IBM and Arm announced a strategic collaboration on April 2, 2026, to enable Arm-based software to run on IBM Z mainframes and LinuxONE systems through virtualization. The partnership aims to bridge the gap between the widely-used Arm AI software ecosystem and IBM's mission-critical enterprise hardware, though no shipping date has been provided.

Technology5 sources

IBM and Arm team up to bring modern AI workloads to enterprise mainframes via virtualization

Technology5 sources

Sharon AI shares surge on $1.8 billion deal despite posting bigger losses in debut earnings

Australia-based Sharon AI announced a $1.8 billion supply agreement and a partnership with Canva, sending shares skyward despite reporting increased losses in its first earnings since its February Nasdaq debut. The artificial intelligence company rents access to high-end computer chips from Nvidia and AMD for AI processing needs.

Business2 sources

Sharon AI shares surge on $1.8 billion deal despite posting bigger losses in debut earnings

Business2 sources

Cognichip raises $60M to use AI for designing the chips that power AI itself

Cognichip raised $60 million led by Seligman Ventures to build AI models that design advanced computer chips. The startup claims its physics-informed platform can reduce chip design costs by over 75% and cut development timelines by more than half. Intel CEO Lip-Bu Tan joined the funding round and will join the board as the company works with over 30 semiconductor firms.

Startups2 sources

Cognichip raises $60M to use AI for designing the chips that power AI itself

Startups2 sources

SpaceX and Google race to build AI data centers in space as scientists question the physics

SpaceX filed to launch up to 1 million satellites for orbital computing, while Google plans its Project Suncatcher test in 2027. The companies promise to solve AI's escalating electricity demands with solar-powered space data centers, but scientists warn of unresolved challenges in thermal management, radiation shielding, and economics that could make the vision unviable for years.

Technology7 sources

SpaceX and Google race to build AI data centers in space as scientists question the physics

Technology7 sources

Supermicro co-founder pleads not guilty to $2.5 billion China chip smuggling charges

Supermicro co-founder Yih-Shyan Wally Liaw entered a not guilty plea to charges of illegally diverting billions in Nvidia AI-powered servers to China. Released on $5 million bond, Liaw faces trial in November alongside two co-defendants in what prosecutors call the highest-profile crackdown on restricted AI technology smuggling. The charges wiped over $6 billion from Supermicro's market value.

Policy5 sources

Supermicro co-founder pleads not guilty to $2.5 billion China chip smuggling charges

Policy5 sources

Nvidia China market share plummets to 55% as Chinese chipmakers capture 41% with 1.65 million AI GPUs

Nvidia's dominance in China has eroded dramatically, with its market share falling from 95% to just 55% as Chinese chipmakers shipped 1.65 million AI GPUs in 2025. Huawei leads domestic vendors with 812,000 units, while U.S. export controls accelerate Beijing's push for self-reliance in semiconductors. The shift marks a critical turning point in the global AI chip race.

Business4 sources

Nvidia China market share plummets to 55% as Chinese chipmakers capture 41% with 1.65 million AI GPUs

Business4 sources

OpenAI closes $122 billion funding round amid fierce AI competition and profitability questions

OpenAI secured $122 billion in the largest Silicon Valley funding round ever, reaching an $852 billion valuation. But the ChatGPT maker faces mounting pressure from rivals like Anthropic and Google while racing toward profitability. The company is pivoting to enterprise revenue and coding tools while planning a unified AI superapp and potential IPO by year-end.

Startups19 sources

OpenAI closes $122 billion funding round amid fierce AI competition and profitability questions

Startups19 sources

Fujitsu and Rapidus partner to build 1.4nm AI chip entirely in Japan with government backing

Fujitsu plans to develop an advanced 1.4nm AI chip manufactured entirely in Japan by Rapidus, marking a significant push toward sovereign AI capabilities. Japan's NEDO will fund roughly two-thirds of the ¥58 billion ($363 million) development cost. The NPU will integrate with Fujitsu's Arm-based Monaka CPUs for deployment in server systems and the Fugaku NEXT supercomputer.

Technology2 sources

Fujitsu and Rapidus partner to build 1.4nm AI chip entirely in Japan with government backing

Technology2 sources

Prev 1…262728…70 Next