TSMC's CoWoS advanced packaging technology, used for AI accelerators with high-bandwidth memory (HBM), faces yield issues and capacity constraints, creating an opportunity for Intel to win foundry cus…
NVIDIA released a white paper at the OCP Global Summit outlining a new AI factory power architecture that shifts to an end-to-end 800 VDC bus, saving up to 26% space versus traditional multi-stage con…
Siemens EDA's Joe Sawicki argues that the semiconductor industry must adopt trustworthy, interpretable AI for chip design and manufacturing, combining deterministic EDA engines with AI to reduce debug…
Nvidia is in early discussions with South Korean inference startup Rebellions for a technical partnership, investment, or acquisition, according to Bloomberg. Rebellions, founded in 2020, has forged c…
Yangtze Memory Technologies Co. Ltd (YMTC), China's sole NAND flash manufacturer, has completed pre-IPO advisory preparations to list on Shanghai's STAR market, following CXMT's IPO that raised nearly…
Large language models (LLMs) consume roughly 20 watts of electricity per query, comparable to a phone charger, whereas the human brain operates on about 20 watts total, a figure refined over 600 milli…
Synopsys has updated its Compute Express Link (CXL) IP portfolio with CXL 4.0 support, targeting AI infrastructure memory bottlenecks. The portfolio integrates a controller, IDE security module, silic…
Arduino's UNO Q board, which combines Debian Linux with the Arduino hardware ecosystem, can run local large language models for embedded applications, according to a practical guide from Arduino. The …
Marvell Technology Inc. introduced a memory-disaggregation portfolio targeting AI infrastructure bottlenecks, including the Bravera SC6 PCIe 6.0 SSD controller, the Structera CXL family, and Photonic …
MIPI Alliance has launched a Physical AI Birds of a Feather group to examine how standardized interfaces can support humanoid robot development, aiming to simplify design, optimize performance, and lo…
Fluid-side observability is emerging as a critical factor in AI hardware reliability as direct liquid cooling becomes standard, according to a sponsored article on EE Times. The article argues that mo…
Intel CEO Lip-Bu Tan signaled a return to memory chip innovation, saying memory is no longer a commodity and that Intel is exploring stacking DRAM on CPUs, amid industry expectations that memory short…
Semiconductor equipment OEMs are turning to build-to-print contract manufacturing to meet rising demand from AI and high-performance computing, avoiding costly in-house expansions. Suppliers operating…
A June Reuters/Ipsos poll found that 57% of Americans would oppose a data center in their community, and 77% were concerned that the AI data center boom would raise electricity prices, as public conse…
NVIDIA Jetson modules can run significantly larger AI workloads than their memory specifications suggest through software-defined memory optimization, according to an EE Times sponsored article. The p…
At DAC 2026, Nvidia's Tim Costa and Synopsys' Ravi Subramaniam highlighted agentic AI, multi-physics, and standards as key themes, with Subramaniam noting that $1.4 trillion is spent on engineering, 9…
CEA-Leti CEO Sébastien Dauvé told the LID World Summit 2026 in Grenoble, held June 23-25, that architecture—not raw compute—is AI's emerging bottleneck, with data movement, interconnects, power conver…
MIT's 2025 GenAI Divide study, analyzing 300 deployments and 150 executive interviews, found that roughly 95% of enterprise generative AI pilots delivered no measurable impact, with only about 5% reac…
Siemens highlights that hardware-assisted verification (HAV) methodologies, including emulation and FPGA-based prototyping, are essential for developing advanced AI chip designs, citing its Veloce CS …
The aerospace and defense sector grew from $884 billion in 2023 to $986 billion in 2024 and is forecast to reach $1.23 trillion by 2028, according to a special report by EE Times that highlights AI-dr…