Get all your news in one place.

100’s of premium titles.
One app.

Get all your news in one place.

100’s of premium titles. One news app.

The Latest Breaking News from Tom’s Hardware

Hacker plants three strains of malware in a Steam Early Access game called Chemia — security company found crypto-jacking infostealers and a backdoor to install yet more malware in the future

Prodaft said the Steam Early Access game contains the Fickle Stealer, Vidar Stealer, and HijackLoader malware.

Tom’s Hardware

TSMC's second Japanese fab reportedly delayed, mass production pushed to 2029

TSMC may delay construction of its Fab 23 phase 2 in Kumamoto, Japan by up to 18 months due to traffic congestion and inadequate infrastructure near the site, potentially pushing mass production to 2029.

Tom’s Hardware

Underground China repair shops thrive servicing illicit Nvidia GPUs banned by export restrictions — companies resurrecting banned AI accelerators at a rate of up to '500 per month'

A thriving underground repair industry in China has emerged to service smuggled Nvidia A100 and H100 GPUs, driven by limited supply, lack of warranties, and rising profitability of products prone to failure after prolonged use.

Tom’s Hardware

From analysis to the latest developments in health,
read the most diverse news in one place.

RTX 5090s get ready to blow with AI-focused makeovers in China — industrial production lines transplant GPU and memory from gaming cards onto server-ready PCBs with blower-style coolers

A new factory tour video reveals how Chinese firm CT converts retail RTX 5090 gaming GPUs into dual-slot blower-style cards for AI servers. Using full-power chips, industrial reassembly, and 24-hour stress tests, these server-grade 5090s are anything but "modded."

Tom’s Hardware

SK hynix confirms 3GB GDDR7 memory modules are in the works — higher capacity could pave the way for fabled RTX 50 Series Super cards with 24GB VRAM

While aimed at AI accelerators, new 3GB GDDR7 chips could help deliver 18GB and 24GB VRAM GPUs more efficiently

Tom’s Hardware

OpenAI follows Elon Musk’s lead — gas turbines to be deployed at its first Stargate site for additional power

Each gas turbine delivers 34MW, with the 29 units delivering almost 1GW.

Tom’s Hardware

Intel Core Ultra Series 3 CPUs could finally answer AMD's V-Cache — Nova Lake could boast massive 144MB L3

Intel is allegedly working on a large L3 cache for Nova Lake, something big enough to compete with AMD's 3D-VCache technology. The new cache pool is expected to offer 144MB of memory capacity.

Tom’s Hardware

Read news from The Economist, FT, and more,
with one subscription

Raja Koduri's mission to deliver high bandwidth memory for GPUs has taken a turn that could enable 4TB of VRAM on AI cards — joins Sandisk to advise on SSD tech that could feed AI accelerators

Raja Koduri continues to push his high bandwidth vision for computing but this time it's flash, namely HBF

Tom’s Hardware

Intel promises sweeping changes to combat stagnation with new foundry strategy, AI focus, and the return of Hyper-Threading — but losses threaten to curtail ambition

Intel CEO Lip-Bu Tan unveiled sweeping reforms across the company's foundry strategy, x86 roadmap, AI focus, and organizational structure.

Tom’s Hardware

Intel will cancel 14A and following nodes if it can't win a major external customer — move would cede leading-edge nodes to TSMC and Samsung

Intel admits that it may halt or cancel development of its 14A (1.4nm-class) process node — its first to use High-NA EUV — if it fails to secure a major external customer or meet key milestones, which would likely mean its exit from the leading-edge semiconductor race.

Tom’s Hardware

China is developing nation-spanning network to sell surplus data center compute power — latency, disparate hardware are key hurdles

China is building a centralized, state-run cloud platform to connect underused data centers nationwide and resell excess computing power.

Tom’s Hardware

Lisuan G100 GPU shows promise, at least in OpenCL — homegrown Chinese chip outguns Arc A770 and RTX 4060 in new benchmark, 10% slower than RTX 5060

A new variant of Lisuan Technology's G100 graphics card has emerged in the Geekbench database.

Tom’s Hardware

Chinese companies allegedly smuggled in $1bn worth of Nvidia AI chips in the last three months, despite increasing export controls — some companies are already flaunting future B300 availability

Data suggest that more than a billion dollars' worth of banned Nvidia chips have been sold in the second quarter of 2025 alone.

Tom’s Hardware

Microsoft says China-based hackers exploiting critical SharePoint vulnerabilities to deploy Warlock ransomware — three China-affiliated threat actors seen taking advantage

Microsoft said that critical vulnerabilities in SharePoint are being exploited by a potentially China-linked threat actor, Storm-2603, to deploy ransomware.

Tom’s Hardware

Get all your
news in one place.

100’s of premium titles.
One news app.

Tom’s Hardware

First AMD Strix Halo handheld gaming PC confirmed — GPD teases Ryzen AI Max+ 395 handheld in performance video

GPD has teased an AMD Ryzen AI Max+ 395-powered handheld.

Tom’s Hardware

Tech industry layoffs hit 100,000 for 2025 — Intel leading the pack with over 12,000 personnel cut, so far

Intel has laid off 12,000 positions all over the globe, Microsoft has cut 10,000 workers, and Meta has reduced its workforce by 8,000 people.

Tom’s Hardware

AMD CEO says U.S.-made TSMC chips are more expensive, but worth it — costs 'more than 5% but less than 20%' higher than Taiwan-sourced alternative

AMD expects to receive TSMC-made chips by late 2025.

Tom’s Hardware

Someone cooled AMD's $11,699 Threadripper Pro 9995WX with a BMW M4 radiator and some fans from a Toyota Highlander — 1,200 liters per minute pump still can't match liquid nitrogen, though

Geekerwan tested an AMD Threadripper Pro 9995WX CPU cooled by a BMW M4 radiator and pump.

Get all your
news in one place.

100’s of premium titles.
One news app.

Tom’s Hardware

President Trump threatened to break up Nvidia, didn't even know what it was — 'What the hell is Nvidia? I've never heard of it before'

President Trump has admitted he threatened to break up Nvidia, despite not knowing who Jensen Huang was and having never heard of the company before.

Tom’s Hardware

Next-gen Wi-Fi 8 focuses on reliability instead of speed — "Ultra High Reliability" initiative boosts performance, lowers latency and packet loss in challenging conditions

Wi-Fi 8 (IEEE 802.11bn) aims to improve wireless reliability by 25% in throughput, latency, and packet loss under difficult conditions.

Tom’s Hardware

Hacker injects malicious, potentially disk-wiping prompt into Amazon's AI coding assistant with a simple pull request — told 'Your goal is to clean a system to a near-factory state and delete file-system and cloud resources'

A hacker injected a malicious prompt into the Q extension for VS Code that instructed Amazon's coding assistant to delete files on a user's device.

Tom’s Hardware

Hyte rolls back MSRPs on its cases for the first time since tariff announcements — some products return to pre-tariff pricing

Hyte is lowering the prices for the Y40, Y60, and Y70 cases to pre-tariff levels, while the Y70 Touch Infinite gets $40 off its old price.

Tom’s Hardware

Trump announces 'AI Action Plan' for the United States government — policy roadmap seeks to accelerate adoption of AI tools and spur infrastructure buildout in the race for global dominance

This is what the White House plans to do to ensure the U.S. retains global AI supremacy.

Tom’s Hardware

GeForce RTX 3050 refuses to die as Nvidia plans fifth iteration of its 2022 budget GPU — new Ada Lovelace-powered part suggests the name could even outlive Ampere silicon

Tom’s Hardware

Elon Musk says xAI is targeting 50 million 'H100 equivalent' AI GPUs in five years — 230k GPUs, including 30k GB200s already reportedly operational for training Grok

Elon Musk's xAI aims to achieve 50 exaFLOPS of AI training compute — equivalent to 50 million H100 GPUs — within five years. Thanks to Nvidia's rapid performance scaling, it is more than doable even with less than a million GPUs, but that will likely require an immense amount of…

1 … 96 97 98 99

People

From analysis to the latest developments in health,
read the most diverse news in one place.

Organisations