Intel and Google published the results of a significant joint security review of Intel Trust Domain Extensions (Intel TDX). The five-month engagement between Intel's INT31 research team and Google Cloud Security focused on hardening Intel TDX Module 1.5, the core software responsible for managing confidential virtual machines. The review identified five vulnerabilities and 35 other weaknesses or areas for improvement. According to Intel, all five vulnerabilities were patched in the latest code version released to partners.

Intel TDX represents a shift in data center security. Unlike traditional virtualization, where the hypervisor has full access to guest data, TDX creates "Trust Domains" (TDs), or hardware-isolated virtual machines. These are cryptographically isolated from the hosting infrastructure using Intel's Secure Arbitration Mode (SEAM). The system relies on the Intel TDX Module, a digitally signed software shim running in SEAM. It enforces security policies that hardware alone cannot handle. Because a compromise of this module could theoretically bypass the system's isolation guarantees, it is a critical verification target.

Get ready to game - the native GeForce NOW app for Linux PCs is now available in beta, letting Linux desktops tap directly into GeForce RTX performance from the cloud. Alongside the expansion comes ten new games, including The Bard's Tale IV: Director's Cut and The Bard's Tale Trilogy for a leveled-up gaming weekend. And operators can squad up in Team Jade's Delta Force when it deploys in the cloud on February 3, bringing high-stakes tactical action to nearly any device with no downloads required.

Linux Just Got RTX
Starting with support for Ubuntu 24.04 and later, the new app delivers a seamless native experience that fits right into Linux desktop workflows. It transforms compatible Linux systems into high-performance GeForce RTX-powered rigs, streaming supported PC titles directly from the cloud at up to 5K resolution and 120 frames per second, or 1080p at 360 FPS. This new desktop-focused app goes beyond the earlier Linux experience that was tuned specifically for Steam Decks, which targeted handheld-friendly resolutions and up to 1200p at 90 FPS. The beta Linux app is built for PCs and notebooks, offering an experience similar to the existing GeForce NOW app on Windows and macOS.

IBM (NYSE: IBM) today announced fourth-quarter 2025 earnings results.

"In the fourth quarter, we delivered strong revenue growth, with double-digit Software performance. Additionally, Infrastructure continued its double-digit revenue growth with the robust adoption of the next generation of our mainframe platform. Our generative AI book of business now stands at more than $12.5 billion. This capped a strong 2025 for IBM where we exceeded expectations for revenue, profit and free cash flow," said Arvind Krishna, IBM chairman, president and chief executive officer. "We enter 2026 with momentum and in a position of strength, giving us confidence in our full-year expectations of more than 5 percent constant currency revenue growth and an increase of about $1 billion in year-over-year free cash flow."

Microsoft just reported its Q4 2025 results, providing some interesting insights into its gaming ventures. The company's overall quarterly revenue increased by an impressive 17% compared to the same period last year, reaching $81.3 billion. This growth is primarily driven by the strong performance of the cloud sector. However, Microsoft's gaming divisions are experiencing challenges with declining revenue. In its Xbox unit, hardware revenue decreased by 32% year-over-year, significantly impacting the business. The overall Xbox division saw a 9% decline, while Xbox content and services dropped by 5%. The services section, which includes the Game Pass subscription, is the only area not experiencing a steep decline. By recently rising Game Pass subscription prices, Microsoft aims to offset some of the hardware losses.

An interesting detail from the earnings report is the Windows OEM and Devices sector, which reported a modest growth of 1%. This is minor considering the Windows 10 end-of-life, which was expected to boost this sector significantly. Windows itself reported a 5% growth, but this is only a slight increase compared to what Microsoft anticipated from OEMs purchasing Windows license keys for Windows 11, as well as a major consumer upgrade spree to Windows 11 devices. With Xbox revenue declining significantly year-over-year for several years, it is becoming less of a priority for the company, which now derives substantial revenue from Microsoft Azure and 365 Commercial subscriptions.

ROG Ally gaming handhelds offer tons of ways to play your favorite games. But cloud streaming continues to up the ante, with Boosteroid boasting high-resolution, high-refresh cloud-streamed games optimized for your ultimate gaming handheld. Boosteroid, the world's biggest independent cloud gaming service, has optimized its infrastructure to better serve ROG Ally series handhelds. Boosteroid has partnered with ROG to make the most of the Ally's architecture for uniquely powerful gaming experiences, including up to 4K, 120 FPS cloud-streamed games.

"We have deepened our strategic cooperation with ASUS to ensure that our software utilizes the capabilities of the ROG Ally series to deliver a highly responsive and visually rich experience that sets a new standard for handheld cloud gaming," said Ivan Shaichenko, CEO of Boosteroid.

North American CSPs' continued investments in AI infrastructure are expected to increase global AI server shipments by more than 28% YoY in 2026, according to the latest market research from TrendForce. The rapid growth of AI inference services is boosting demand for general-purpose servers, supporting both replacement and expansion efforts. Consequently, TrendForce predicts that total global server shipments, including AI servers, will accelerate from 2025, with a 12.8% YoY growth in 2026.

The server market from 2024 to 2025 primarily centered on training advanced large language models (LLMs), utilizing AI servers with GPUs and HBM for parallel computing. However, starting in the second half of 2025, the growth of AI agents, LLaMA-based applications, and Copilot upgrades shifted CSP's focus toward inference services as a monetization strategy. AI inference workloads are now deployed not only on dedicated AI server racks but also on general-purpose servers that handle pre- and post-inference computing and storage.

Hard Disk Drives are now in great demand, which has pushed their price premiums up by an average of 46% since mid-September. According to the ComputerBase tracker, HDDs have increased their prices over the past four months by anywhere from 23% to 66%, depending on the model. Standing at the lowest 23% increase is the Seagate IronWolf Pro NAS HDD + Rescue 16 TB HDD, while the highest price increase was observed in the Toshiba Cloud Scale Capacity MG10F AFA 22 TB model. The price tracker followed fluctuations of the 12 most popular HDD models based on their EUR listings and found that HDDs rose by an average of 46% across the four-month period, resulting in spinning rust being almost 50% more expensive, without any substantial supply chain shortage.

A month ago in December, DigiTimes reported that fourth-quarter 2025 contract negotiations concluded with traditional HDD prices rising about 4% quarter-over-quarter, the largest increase in the past eight quarters. This contributed to the overall 46% price increase observed today, as these single-digit contract changes accumulate and multiply by the time they reach consumers. Interestingly, HDDs contain almost no silicon for storage purposes, making their significant price increases a supply chain issue unrelated to the semiconductor industry. Apart from the controllers that use silicon, HDD platters are made from materials that are not currently in short supply. However, high demand is what keeps their prices elevated.

EnGenius Technologies Inc., a global leader in advanced connectivity and cloud-managed networking solutions, today announces the expansion of its EnGenius Cloud-managed security portfolio with advanced Wi-Fi 7 access points featuring EnGenius AirGuard, now integrated into its flagship ECW536S Wi-Fi 7 series for enhanced protection in security-sensitive environments such as financial institutions, healthcare, and enterprises.

Continuous Wireless Threat Protection with AirGuard
As enterprise environments embrace remote access and proliferating Bluetooth/IoT endpoints, attack surfaces have expanded dramatically. EnGenius addresses this challenge with AirGuard, an intelligent wireless intrusion detection and prevention system (WIDS/WIPS) built into its APs. AirGuard delivers 24/7 threat detection—identifying and neutralizing threats such as evil twins, rogue APs, man-in-the-middle attacks, RF jammers, and flood attempts—by leveraging dedicated scanning radios. This always-on security protection operates without compromising wireless performance. Professional-grade RF spectrum analysis visualizes SSID legitimacy and ensures optimal channel utilization, while zero-wait DFS facilitates seamless, non-disruptive channel shifts when radar is detected. Additionally, BLE scanning capabilities detect nearby Bluetooth devices.

Mobvoi, a global leader in AI interaction and hardware, and Olares, the innovator behind the Personal AI Cloud, today announced a partnership designed to redefine the modern workflow. By integrating Mobvoi's TicNote Pods with the computational power of Olares One, the two companies are introducing a flexible ecosystem that offers users an unprecedented choice between subscription-based cloud services and private, local processing.

The partnership centers on the TicNote Pods, a device that has already garnered significant industry acclaim as the world's first 4G-enabled AI note-taking earbuds. Unlike traditional wearables that rely on a tethered smartphone, TicNote Pods operate autonomously. Whether in a boardroom, a lecture hall, or on a run, these earbuds capture high-fidelity audio and instant voice memos, ensuring that no critical information or sudden flash of inspiration is ever lost.

The cloud accelerator market share study was conducted at UBS bank, one of the leading research hubs and close followers of market trends. According to UBS, NVIDIA commands 71.2% of the cloud accelerator location share, which includes GPUs and other types of ASICs. The study also included AMD, and other ASICs. For the AMD front, the situation is entirely different as the data shows AMD only captures 5.8% of the cloud accelerator location share. The remaining portion is split between ASICs designed by the likes of AWS, Meta, Broadcom, MediaTek, and many others, which command 22.3%.

For NVIDIA, the company has managed to penetrate its GPUs in 258 datacenter locations, serving as cloud GPU providers. This is a significant number considering that some of these customers are ordering tens and even hundreds of thousands of GPUs at once, especially as the demand expands. NVIDIA is still the backbone of the entire AI infrastructure, and even its older GPUs, which include A100 and H100, are capturing a significant part of the GPU cloud market share. As "Blackwell" SKUs, designed for a million GPU scale systems, are scaled, and we enter the "Rubin" era, NVIDIA's lead may only expand.

D-Link Corporation, a global leader in networking and connectivity solutions, announced the launch of Nuclias Unity, a next-generation license-free cloud network management platform. Purpose-built for organizations ranging from SMBs to large, multi-site enterprises, Nuclias Unity delivers unified control, simplified operations, and enterprise-grade reliability—without the traditional cost and complexity of licensed cloud platforms.

Backed by D-Link's decades of networking expertise, Nuclias Unity empowers IT teams to manage wired and wireless networks through a unified cloud management platform, accelerating cloud adoption while ensuring visibility, consistency, and operational efficiency across all business environments.

The world's top-performing system for graph processing at scale was built on a commercially available cluster. NVIDIA last month announced a record-breaking benchmark result of 410 trillion traversed edges per second (TEPS), ranking No. 1 on the 31st Graph500 breadth-first search (BFS) list. Performed on an accelerated computing cluster hosted in a CoreWeave data center in Dallas, the winning run used 8,192 NVIDIA H100 GPUs to process a graph with 2.2 trillion vertices and 35 trillion edges. This result is more than double the performance of comparable solutions on the list, including those hosted in national labs.

To put this performance in perspective, say every person on Earth has 150 friends. This would represent 1.2 trillion edges in a graph of social relationships. The level of performance recently achieved by NVIDIA and CoreWeave enables searching through every friend relationship on Earth in just about three milliseconds. Speed at that scale is half the story—the real breakthrough is efficiency. A comparable entry in the top 10 runs of the Graph500 list used about 9,000 nodes, while the winning run from NVIDIA used just over 1,000 nodes, delivering 3x better performance per dollar.

Lenovo today announced an expansive series of new data storage, virtualization solutions and data management services, designed to help customers modernize their IT and data infrastructure for powering enterprise applications and AI ready capabilities. Today's new offerings include new Lenovo ThinkSystem and ThinkAgile data storage and virtualization solutions, announced in tandem with data management services. Designed to provide a modern foundation for enterprises and mid-size businesses achieving AI innovations, this announcement combines complementary hardware, software and services offerings to help deploy, manage, and unleash the true potential of enterprise data.

"Sixty-three percent of organizations either do not have or are unsure if they have the right data management practices for AI, according to a survey by Gartner". Customer unique data is the differentiation that will drive their competitive advantage and most accurate results, however 80% of storage deployed in the last 5 years is on slower hard drive-based storage (according to IDC reporting) that is not optimized for AI. At the same time, customers are challenged with new virtualization and containerization requirements that demand open solutions. Businesses must mitigate this risk by ensuring their enterprise data systems and practices are modernized for advanced use cases.

IBM (NYSE: IBM) and Confluent, Inc. (NASDAQ:CFLT), the data streaming pioneer, today announced they have entered into a definitive agreement under which IBM will acquire all of the issued and outstanding common shares of Confluent for $31 per share, representing an enterprise value of $11 billion. Confluent provides a leading open-source enterprise data streaming platform that connects, processes and governs reusable and reliable data and events in real time, foundational for the deployment of AI.

IDC estimates that more than one billion new logical applications will emerge by 2028, reshaping technology architectures across industries. To fuel meaningful outcomes and drive productivity in operations, these applications, as well as AI agents, need access to connected and trusted data - in real time. IBM and Confluent will enable end-to-end integration of applications, analytics, data systems and AI agents to drive intelligence and resilience in hybrid cloud environments.

Scaleway, Europe's leading cloud provider, and Fujitsu, a global leader in high performance computing, today announced a strategic collaboration to explore CPU-based AI inference as a sustainable and sovereign option for European organizations. By combining Fujitsu's advanced processor research with Scaleway's European cloud expertise, the two companies aim to evaluate new ways to support efficient, responsible and secure AI deployment at scale.

Europe's rapid adoption of AI has increased demand for infrastructures that reconcile performance, energy efficiency and data sovereignty. Many organizations now operate AI models continuously in production and require predictable performance, controlled operating costs and a reduced environmental footprint. CPU-based architectures are particularly well suited to these requirements as they offer stable performance, efficient energy use and simple integration within existing environments. MONAKA has been designed for this purpose and to complement the AI systems already used within Europe.

Serafim, a leader in mobile and cloud gaming hardware innovation, today announced its upcoming CES 2026 exhibition, where the company will debut its latest flagship devices: the Serafim R2+ Racing Wheel and the Serafim S3 Cloud Gaming Controller. As gaming continues to evolve across mobile, cloud-based, and cross-platform ecosystems, Serafim remains committed to developing advanced solutions that enhance player experience worldwide.

Introducing the Serafim R2+ Racing Wheel
Designed for modern gamers who demand flexibility and realism, the R2+ Racing Wheel delivers console-grade performance across mobile and traditional gaming platforms.

In November, we have several updates coming your way as we look to continue improving your gaming experiences on the devices which you like to play. Starting with Gaming Copilot (Beta), this feature is now available on the Xbox mobile app, adding your very own personal gaming sidekick directly to the phone in your pocket. Full screen experience (FSE), first seen on the ROG Xbox Ally, is now available to all players on even more Windows 11 handhelds. It's also expanding to a wider range of Windows 11 devices through the Xbox Insider and Windows Insider programs. In cloud gaming news, players who stream their games via cloud gaming can now choose their streaming resolution for select titles, with support for up to 1440p resolution.

Xbox Cloud Gaming is now available in even more countries, including India, and we've also brought support for additional devices in Brazil and Argentina. Plus, the Xbox app is coming soon for the Amazon Fire TV Stick 4K (2025), allowing you to stream Xbox games directly to your favorite device while opening up more ways for our players around the world to play. Read about these and much more, including new accessories and game additions, below.

Microsoft has acknowledged that File Explorer on Windows 11 has performance issues, which is concerning for such a crucial application within an operating system. To address this, Microsoft is considering preloading File Explorer in the background to enable quicker access when users open the application. In the Windows 11 Insider Preview Build 26220.7271 (KB5070307) available in the Dev & Beta channels, Microsoft stated, "We're exploring preloading File Explorer in the background to help improve File Explorer launch performance. This shouldn't be visible to you, outside of File Explorer hopefully launching faster when you need to use it."

If you prefer the traditional behavior, you can disable window preloading. Open File Explorer, navigate to Folder Options, select the View tab, and uncheck Enable window preloading for faster launch times. Microsoft is seeking feedback, and if you encounter any issues, you can report them in the Feedback Hub under Files, Folders and Online Storage > File Explorer Performance or Files, Folders and Online Storage > File Explorer.

We are approaching the release of version 1.0 of the renowned Microsoft PowerToys with the v0.96 update. This latest version, PowerToys 0.96, is a substantial update that focuses on enhancing intelligent clipboard operations. Microsoft has completely revamped the Advanced Paste feature, moving away from exclusive provider lock-in and offering users the flexibility to connect with either cloud-based AI services or local inference engines. The redesigned Advanced Paste now supports a wider range of content formats, including markdown, JSON, HTML fragments, images, and multimedia files, with improved processing capabilities.

Optical character recognition allows for text extraction from images, while integrated transcoding capabilities convert audio and video into standard formats when needed. Users can configure multiple endpoints simultaneously from major platforms such as Azure OpenAI, OpenAI's flagship models, Google's Gemini, and Mistral's offerings. Additionally, operations can be directed to private infrastructure through local solutions like Foundry Local or Ollama, utilizing onboard neural processing units and GPUs for offline execution.

Western Digital has quietly introduced a new ultra-dense storage rack called the Ultrastar Data102. This 4U rack shelf holds 102 3.5-inch drives, offering just over three petabytes of raw storage when equipped with today's 30 TB hard disks. Upgrading to 32 TB drives increases the raw capacity to about 3.26 PB, providing a modest but valuable boost for customers aiming to maximize terabytes per rack unit. The chassis is designed for dense, continuous operation, featuring redundant, hot-swappable power supplies, front-serviceable cooling, and SAS host connectivity suitable for large storage fabrics.

Western Digital's product documentation indicates 102 drive bays without separate SSD mounting points. However, there is room for up to 3.84 TB of SSD capacity for a hybrid storage design. A fully populated enclosure is both heavy and power-intensive, accommodating up to two 80 Plus Platinum 1800 W power supplies. Cloud providers, archival services, and AI training clusters that prioritize raw capacity and cost per TB will find the Ultrastar Data102 appealing. However, organizations needing integrated flash tiers might opt for enclosures that explicitly include SSD bays. As with other enterprise-grade systems, the final price is typically negotiated through OEM and integrator channels rather than listed as a retail figure.

Chaos has entered the chat. It's GFN Thursday, and things are getting intense with the launch of Call of Duty: Black Ops 7, streaming at launch this week on GeForce NOW. Dive straight into the chaos across devices - underpowered laptops, Macs, and Steam Decks. Power up the week with 12 new games to stream in the cloud, including the latest GeForce RTX 5080-optimized game, Anno 117: Pax Romana. Phoenix is the latest region to get GeForce RTX 5080-class power, with Stockholm coming up next. Stay tuned to GFN Thursday for updates as more regions upgrade to Blackwell RTX. Follow along with the latest progress on the server rollout page.

The Biggest "Black Ops" Ever
In Call of Duty: Black Ops 7, Treyarch and Raven Software are bringing players the biggest Black Ops ever. The year is 2035, and the world teeters on the edge of collapse after the global events following Black Ops 2 and Black Ops 6. Wielding cutting-edge technology, the Black Ops team led by David Mason must fight back against a manipulative enemy who weaponizes fear above all else.

Olares, a technology startup dedicated to open and local AI, today announced the upcoming launch of its debut product, Olares One, a personal AI cloud that delivers cloud-level AI performance with local processing and data ownership by design. The company is backed by a $45 million Series A funding round led by SIG. Engineered for AI enthusiasts, content creators, developers, and privacy-conscious businesses, Olares One packs server-grade strength into a whisper-quiet desktop form factor. It features an NVIDIA GeForce RTX 5090 Mobile GPU (24 GB GDDR7), an Intel Core Ultra 9 275HX processor, and 96 GB of RAM.

"Our goal is to return control of AI to the individual," said Peng Peng, Founder of Olares. "With Olares One, we are providing the powerful hardware and open software needed to run advanced AI models locally. This empowers users to leverage the benefits of AI with the confidence that their privacy is protected."

Arm (NASDAQ: ARM) published a letter to its shareholders containing results for the second quarter of fiscal year 2026 ended September 30, 2025. The infographic below offers the key highlights.

Here are what the results mean and how they were achieved:

• Surpassing $1 billion for the third consecutive quarter, as Arm reported second-quarter revenue of $1.14 billion, up 34% year-over-year. Revenue was above the high end of guidance, and non-GAAP EPS finished above the high end.
• Royalty revenue rose 21% year-over-year to $620 million, driven by the continued adoption of Arm technology with higher royalty rates per chip, such as Armv9 and Arm Compute Subsystems (CSS), and increased usage of Arm-based chips in data centers.
• Three new Arm CSS licenses signed in the quarter, one each in smartphones, tablet, and data centers, bringing the total to 19 licenses with 11 companies, with five customers already shipping CSS-based chips. Samsung will also leverage CSS for its Exynos family of chipsets, which means the top four Android phone vendors are now shipping CSS-powered devices.
  • In Q2, Arm launched the Lumex CSS platform, our most advanced smartphone platform to date, delivering up to 5x faster AI performance on a new generation of Arm C1 CPUs with Arm Scalable Matrix Extension 2 (SME2) and up to 3x the energy savings, alongside up to 2x AI and graphics performance through the integration with the new Arm Mali G1-Ultra GPU compared to the prior generations. The ecosystem is already embracing Lumex, with leading applications, such as Alipay, Gmail, and YouTube, accelerating on-device inference and real-time personalization, and partners, such as MediaTek, designing Lumex configurations into its next-generation chips for new flagship smartphones from Oppo and vivo.
  
  Read full story

AI factories continue to grow at unprecedented scale, processing structured, unstructured and emerging AI-native data. With demand for trillion-token workloads exploding, a new class of infrastructure is required to keep pace. At NVIDIA GTC Washington, D.C, NVIDIA revealed the NVIDIA BlueField-4 data processing unit, part of the full-stack BlueField platform that accelerates gigascale AI infrastructure, delivering massive computing performance, supporting 800 Gb/s of throughput and enabling high-performance inference processing.

Powered by software-defined acceleration across AI data storage, networking and security, NVIDIA BlueField-4 transforms data centers into secure, intelligent AI infrastructure—designed to accelerate every workload, in every AI factory. It's purpose-built as the end-to-end engine for a new class of AI storage platforms, bringing AI data storage acceleration to the foundation of AI data pipelines for efficient data processing and breakthrough performance at scale.

Supermicro, Inc., a Total IT Solution Provider for AI, Cloud, Storage, and 5G/Edge, is announcing the latest addition to the AMD based EPYC 4005 series CPU Servers, optimized to deliver powerful performance, efficiency, and affordability. Supermicro will feature the new MicroBlade multi-node solution, ideal for companies looking for a scalable and cost-effective solution with centralized management for power, cooling, and networking.

"We continue to expand our portfolio with the new Supermicro MicroBlade, which incorporates AMD EPYC 4005 series processors," says Charles Liang, president and CEO of Supermicro. "This addition to our successful EPYC-based product line provides a very cost-effective, green computing solution for Cloud Service Providers. A single 48U rack can have up to 160 servers and 16 Ethernet switches, enabling tremendous computing power with built-in switches per enclosure."