AI Compute

Twitter

Facebook

Copy Link

Key Facts

Key PlayersNVIDIA, Microsoft, Google, AWS, CoreWeave

Top News

Sandisk Stock Rockets 22% as AI Memory Boom Fuels 830% Rally and Price Surge

Sandisk stock soared 22% to $334.75, extending an 830% rally since its February 2025 spin-off. Memory prices are projected to jump 40-70% in Q1 2026 as AI-driven demand for enterprise SSDs and NAND flash storage outpaces supply. Nvidia's new storage platform for agentic AI inference adds fuel to the memory supercycle.

Business2 sources

Sandisk Stock Rockets 22% as AI Memory Boom Fuels 830% Rally and Price Surge

Business2 sources

Microsoft partners with MISO to deploy AI for grid modernization across 15 U.S. states

Microsoft has partnered with the Midcontinent Independent System Operator to modernize the Midwest power system serving 42 million people across 15 states. The collaboration will deploy Microsoft Azure and Foundry AI to predict weather disruptions, optimize transmission line planning, and accelerate grid operations as AI power demand from data centers pushes electricity needs to record highs.

Technology4 sources

Microsoft partners with MISO to deploy AI for grid modernization across 15 U.S. states

Technology4 sources

AI Demand Triggers Global Memory Shortage as RAM Prices Surge Up to 70% in 2026

An unprecedented memory chip shortage is reshaping the tech industry as Samsung, SK Hynix, and Micron prioritize AI data centers over consumer devices. DRAM prices are jumping 50-70% in Q1 2026 alone, driving record profits for memory makers while PC manufacturers and consumers face sharp price increases. The shift from consumer to AI-focused production could permanently alter the memory market landscape.

Business19 sources

AI Demand Triggers Global Memory Shortage as RAM Prices Surge Up to 70% in 2026

Business19 sources

Nvidia unveils DLSS 4.5 with 6x Frame Generation and improved AI upscaling at CES 2026

Nvidia introduced DLSS 4.5 at CES 2026, featuring a second-generation transformer model that delivers improved image quality across all RTX GPUs. The update includes a new 6x Multi Frame Generation mode for RTX 50-series cards that generates five additional frames for every rendered frame, targeting 240Hz 4K gaming on ultra-high-refresh-rate monitors.

Technology19 sources

Nvidia unveils DLSS 4.5 with 6x Frame Generation and improved AI upscaling at CES 2026

Technology19 sources

AMD unveils Helios rack-scale AI system with 72 MI455X accelerators and 256-core EPYC Venice

AMD revealed its Helios rack-scale AI system at CES, featuring 72 Instinct MI455X accelerators with 31 TB of HBM4 memory and delivering 2.9 FP4 exaflops for AI inference. The system pairs with AMD's 256-core EPYC Venice CPUs and targets hyperscalers like OpenAI, xAI, and Meta. AMD also introduced the MI400-series family, including MI430X for HPC and MI440X for enterprise deployments.

Technology7 sources

AMD unveils Helios rack-scale AI system with 72 MI455X accelerators and 256-core EPYC Venice

Technology7 sources

Nvidia Vera Rubin architecture slashes AI costs by 10x with advanced networking at its core

Nvidia CEO Jensen Huang announced the Vera Rubin architecture at CES 2026, declaring it's in full production. The next-generation AI superchip platform promises a 10x reduction in inference costs and requires four times fewer GPUs to train certain models compared to Blackwell. But the real innovation lies in its six-chip design, where advanced networking components work in concert to handle distributed AI workloads across data centers.

Technology32 sources

Nvidia Vera Rubin architecture slashes AI costs by 10x with advanced networking at its core

Technology32 sources

FuriosaAI launches RNGD chip in mass production to challenge Nvidia's AI inference dominance

South Korean startup FuriosaAI has begun mass production of its RNGD neural processing chip, designed specifically for AI inference workloads. Founded by former Samsung engineer June Paik, the company claims its chip delivers up to 2.25 times better inference performance than competing GPUs while consuming less power. Valued at over $700 million after rejecting an $800 million acquisition offer from Meta, FuriosaAI aims to break Nvidia's market dominance.

Technology2 sources

FuriosaAI launches RNGD chip in mass production to challenge Nvidia's AI inference dominance

Technology2 sources

MicroCloud Hologram unveils quantum spectral filter tech for hybrid graph neural networks

MicroCloud Hologram Inc. released learnable quantum spectral filter technology that fuses quantum computing with graph neural networks. The breakthrough enables networks with one million nodes to run on just 20 qubits, offering exponential compression for large-scale graph learning in social media, traffic networks, and internet connectivity applications.

Technology2 sources

MicroCloud Hologram unveils quantum spectral filter tech for hybrid graph neural networks

Technology2 sources

Nvidia's $20B Groq bet and Vera Rubin platform reveal how AI inference is splitting the GPU era

Nvidia CEO Jensen Huang outlined a major strategic shift at CES 2026, emphasizing serviceability and inference economics while announcing a $20 billion Groq licensing deal. The move signals the end of one-size-fits-all GPUs as AI inference workloads split into prefill and decode phases, with the Vera Rubin platform designed for modular maintenance and continuous operation in constrained power environments.

Technology3 sources

Nvidia's $20B Groq bet and Vera Rubin platform reveal how AI inference is splitting the GPU era

Technology3 sources

Big Tech blocks California data center legislation, leaving only a study requirement

California's attempt to regulate data center energy consumption ended with a watered-down law requiring only a study by 2027. Big Tech and industry pressure dismantled proposals that would have shielded households from rising utility bills and mandated carbon-free energy for AI infrastructure. Lawmakers plan to revive tougher regulations in 2026.

Policy2 sources

Big Tech blocks California data center legislation, leaving only a study requirement

Policy2 sources

Investors warn AI bubble poses major economic risk as inflation fears and unsustainable spending mount

Global investors are sounding alarms about an AI bubble threatening markets in 2026, as massive investments in artificial intelligence drive inflation concerns. With $400 billion spent in 2025 and data center costs spiraling, money managers question whether AI's inflated valuations can sustain themselves amid rising chip and energy costs.

Business6 sources

Investors warn AI bubble poses major economic risk as inflation fears and unsustainable spending mount

Business6 sources

WiMi Hologram stock surges as quantum computing breakthrough promises practical AI applications

WiMi Hologram Cloud Inc. shares jumped after announcing two quantum computing breakthroughs that integrate quantum modules into classical deep learning systems. The AR technology provider's MC-QCNN and QB-Net innovations reduce parameters by up to 30 times while maintaining performance, marking a shift from theoretical quantum AI toward commercial implementation in image classification, medical imaging, and video analysis applications.

Technology3 sources

WiMi Hologram stock surges as quantum computing breakthrough promises practical AI applications

Technology3 sources

Nvidia skips gaming GPUs at CES 2026, doubles down on AI as shortages squeeze PC gamers

For the first time in five years, Nvidia declined to announce new gaming GPUs at CES 2026, focusing instead on its AI business. CEO Jensen Huang spent nearly two hours discussing the Rubin platform for AI data centers while memory chip shortages drive GPU prices skyward. The RTX 5090 now fetches $4,000 at some retailers, and Nvidia may bring back the 2021 RTX 3060 to address supply constraints.

Technology14 sources

Nvidia skips gaming GPUs at CES 2026, doubles down on AI as shortages squeeze PC gamers

Technology14 sources

India positions itself as global AI development hub with Impact Summit 2026 and startup push

Prime Minister Narendra Modi met with 12 Indian AI startups ahead of the India AI Impact Summit 2026, calling them 'co-architects of India's future.' The summit, scheduled for February 15-20 in Delhi, will bring together leaders from 20 countries and showcase India's growing AI ambitions through case studies, compendiums, and homegrown solutions focused on ethical and inclusive AI development.

Policy7 sources

India positions itself as global AI development hub with Impact Summit 2026 and startup push

Policy7 sources

DeepSeek unveils mHC architecture that could reshape how developers train advanced AI models

Chinese AI firm DeepSeek has released research on Manifold-Constrained Hyper-Connections (mHC), a new AI training method that addresses signal degradation in neural networks. The framework could enable smaller developers to build frontier models without massive computational costs, potentially previewing the architecture behind the anticipated DeepSeek R2 model expected around February's Spring Festival.

Science7 sources

DeepSeek unveils mHC architecture that could reshape how developers train advanced AI models

Science7 sources

Prev 1…596061…78 Next