Arm has hired Amazon's director of engineering for Amazon Web Services, who helped develop its homegrown AI hardware. This is part of ARM's drive towards developing its own complete processors, rather than just designing them for third parties.
China's push for AI hardware self-reliance hits the roadblock despite U.S. government's restructions due to fragmented hardware and software ecosystems.
Experts say that China has no power supply issues, allowing it to build as many data centers as it wants without worrying about exceeding its grid's capacity.
The rapid growth of AI data centers is driving up U.S. household power bills as utilities expand the grid to meet their massive, volatile electricity demand.
Estimates from the University of Rhode Island's AI lab suggest OpenAI's more capable GPT 5 could consume up to 45 GWh per day. But researchers may be wrong.
OpenAI is facing backlash over its introduction of the new GPT-5 model, while removing older options. Users claim it's significantly worse, despite it achieving greater scores on a range of benchmarks.
Huawei's newest software tool, called the Unified Cache Manager, seeks to optimize utilization across the traditional cache hierarchy for AI inference, allowing China's AI firms to be more competitive without the need for exotic HBM memory that is near-impossible to obtain in the country.
Apple’s Siri AI test reportedly hit a snag as the company imposed strict limits on banking, financial and health apps to avoid repeating embarrassing mistakes.
Microsoft is putting Copilot Gaming (Beta) on the Game Bar for Xbox Insiders, while giving every creator access to its new AI-driven 2D to 3D feature (as long as you have a Microsoft Account).
Tesla is shutting down its Dojo supercomputer program, reassigning staff and leaning more on AMD and Nvidia. Tesla's eventual designs expected to rely on a unified architecture spanning from edge devices to data centers.
Original Apple employee, Daniel Kottke, was spotted at a recent Computer History Museum event where he was demoing a unique project: An original Apple 1 replica with a modern Wi-Fi module, running a version of ChatGPT.
Lagging behind leading tech firms in artificial intelligence, Apple is ramping up internal efforts and actively pursuing acquisitions to accelerate its AI development, but at the same time keeping CapEx modest compared to peers like Google, Meta, and Microsoft.
Microsoft reached a $4 trillion valuation on strong fiscal 2025 results, surging cloud and AI growth, deep ties with OpenAI, its own AI development, and rising demand ahead of Windows 10's end-of-support.
Meta CEO Mark Zuckerberg outlines his vision for “personal superintelligence” just ahead of the company’s Q2 earnings. With billions poured into AI hires and hardware partnerships, Meta's new direction hints at a future shaped by open-source models and personalized AI infrastructure.
The European Union is launching a $30 billion initiative to build a network of gigawatt-class AI data centers across 16 member states in a bid to catch up with the U.S. and China, though concerns remain over power supply, unclear specifications, and long-term business viability.
A Redditor has demonstrated the ability of the recently launched ChatGPT agent to casually swat away a captcha, so that it can complete its assigned task(s).
Chinese AI hardware and software developers form alliances to develop AI standards to compete against American AI technologies and deploy AI across a broad set of applications.
Cloudflare is testing Positron AI's Atlas machine based on Archer accelerators, an inference-only solution that claims to outperform Nvidia's H200 DGX using one-third the power.
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 power.
OpenAI and Oracle plan to add 4.5 GW of U.S. data center power for the Stargate project, boosting total capacity past 5 GW to support millions of AI processors.
AMD and Stability AI have launched the first Stable Diffusion 3.0 Medium model optimized for BF16 and XDNA 2 NPUs, enabling high-quality, offline image generation on Ryzen AI laptops.
When you update your iPhone to iOS 26 and turn on Apple Intelligence, previously unavailable notification summaries for news apps will be turned on automatically. The fourth beta of Apple’s upcoming iOS 26 operating system has re-enabled the Apple Intelligence-powered...
Przemysław “Psyho” Dębiak, a Polish programmer and former OpenAI engineer, defeated the company’s own advanced AI model in the 2025 AtCoder World Tour Finals. In a grueling 10-hour coding contest, he outscored the AI by 9.5%, proving human ingenuity still has an edge, at least for now.
登录
