Category
TeamT5 reports that attack volume from a Chinese state-linked group it tracks more than doubled after adopting AI. Drawing on public case studies, this piece examines how AI now links reconnaissance, code generation, and lateral movement to speed up attack workflows—and what conditions defenders need to monitor.
Micron says HBM3E uses roughly three times more wafer capacity than DDR5 at the same process node, and that this input ratio will rise further in future generations. Here's how expanding bandwidth demands are straining DRAM supply and thermal design.
Maxell has developed a wireless-charging solid-state battery module in a 1/2AA form factor rated up to 150°C, aimed at food temperature loggers. Practical validation with food maker House Foods is still underway.
Amazon has signed a definitive agreement to acquire DuckLabs. Here's how DuckDB's intellectual property remains with a foundation, how the technical direction may shift as core developers join AWS, and what it means for the project's neutrality going forward.
Meta has agreed to an $18 billion (~¥2.86 trillion) settlement with attorneys general from 52 states, territories, and the District of Columbia over teen protection. Thirty percent of the payment is contingent on TikTok and YouTube adopting similar measures, revealing a negotiating strategy aimed at turning Meta's own standards into industry norms. The company expects to record roughly $10 billion in related costs in Q3.
NVIDIA's Q2 earnings, released August 26, 2026, showed revenue of $96.2 billion, beating market expectations. The unusual reversal in which GAAP net income exceeded Non-GAAP net income was driven by $7.7 billion in net equity securities gains, not GPU sales. The Q3 guidance of $108 billion—which would top the $96.2 billion actual result—still assumes zero data center revenue from China.
NVIDIA has announced NVHBM, which moves HBM control circuitry to the base die. The company claims gains in bandwidth, power, and compute area, but absolute specifications, mass-production timing, and adopting products remain undisclosed.
The anonymous model Ox Alpha turns out to be Z.ai's GLM-5.3-Flash. Here's how its inference was designed to run on tens of thousands of Chinese-made AI chips, and why the claimed 100 trillion tokens-per-day capacity figure needs context.
Fujitsu's 144-core MONAKA 3D-stacks N2P compute cores with N5 SRAM and I/O. Ahead of 2027 shipment, estimated performance and stacking still need real-world verification.
TrendForce forecasts that DRAM and NAND will account for 68% of major cloud providers' capital expenditure in 2027. We examine the assumptions behind this estimate, the role of long-term contracts, and the impact of reduced memory configurations.
