SSPAI

Category: NEWS

  • SSPAI Morning Brief: Huawei Launches New Phones, Tablets and Smart Devices at Spring Event

    SSPAI Morning Brief: Huawei Launches New Phones, Tablets and Smart Devices at Spring Event

    Morning Brief

    1. Huawei holds spring all-scenario product launch event
    2. DJI sues Insta360 over six patent ownership disputes
    3. Douyin releases “Douyin Online Abuse Governance Guidelines (2026 Edition)”
    4. News Worth a Quick Look

    Huawei holds spring all-scenario product launch event

    On March 23, Huawei held its spring all-scenario product launch event in Changsha, unveiling more than a dozen new products across multiple categories.

    The HUAWEI Mate 80 Pro Max Fengchi Edition introduces a new cooling structure integrated into the camera module, featuring a hidden, seamless airflow design. It is the first to support both 90 FPS frame rates and ray tracing with balanced image quality enabled simultaneously. Overall performance is improved by an average of 45% compared to the Mate 70 Pro+, while HyperSpace Memory technology boosts background app retention by 67%. The device shares similar specifications with the Mate 80 Pro Max, including the Kirin 9030 Pro processor and 16GB RAM, but replaces the 50MP periscope telephoto lens with a triple-camera setup plus a second-generation Red Maple color camera. It will go on sale at 10:08 on March 27, priced at RMB 8,499 for 16GB+512GB and RMB 9,499 for 16GB+1TB.

    Huawei also announced the Enjoy 90 series, including the Enjoy 90, Enjoy 90 Plus, and Enjoy 90 Pro Max. The series features Huawei’s “Giant Whale” battery, HarmonyOS smart communication, Wi-Fi 7 support, and a “Xuanjia” durable body design. The Pro Max and Plus models are powered by the Kirin 8000 chipset, while the base model uses the Kirin 8000A. Battery capacities are 6620 mAh for the Enjoy 90 and 90 Plus, and 8500 mAh for the Pro Max. The Enjoy 90 and 90 Plus feature 6.67-inch 120Hz LCD displays, while the Pro Max upgrades to a 6.84-inch 120Hz OLED panel. The Pro Max and Plus models are available for pre-order starting March 23 at RMB 1,499, while the base model opens pre-orders on April 23 starting at RMB 1,299.

    The HUAWEI WATCH Ultimate 2 adds a new Wilderness Green color. Its golf features have been upgraded with practice range and course modes, including club switching, swing monitoring, customizable flag positions, rotating green views, and a new 18-hole bezel scorecard, enhancing the overall golf experience. The new color variant is priced at RMB 7,999.

    The HUAWEI WATCH GT Runner 2 is designed for runners, featuring a suspended antenna architecture that improves signal strength by 250% and enhances positioning accuracy by 20% compared to the GT 6 series. Built with a titanium alloy body, it weighs 43.5g and offers up to 32 hours of outdoor battery life. It comes in three colors—Blue, Orange, and Black—and is priced at RMB 2,588, with sales starting April 10.

    The Huawei Band 11 series includes the Band 11 Pro and Band 11. Both feature a 1.62-inch display, with the Pro version reaching a peak brightness of 2000 nits. The body is as thin as 8.99mm, with the Pro model using aluminum alloy and available in three colors. The standard model offers both aluminum alloy and reinforced polyester fiber options. The series includes sleep apnea screening, all-day HRV monitoring, and 12 emotional state tracking modes. The Pro model also supports standalone GNSS positioning. Prices start at RMB 389 for the Pro and RMB 269 for the standard version.

    The Huawei MatePad Pro 12.2-inch features a dual-layer OLED soft-light display with a 3:2 aspect ratio, 12GB RAM, a 10,000 mAh battery, Wi-Fi 7 support, and a slim 5.5mm body weighing 508g. It is available in four colors and supports HarmonyOS desktop apps. Pricing starts at RMB 4,499, with sales beginning March 28.

    The Huawei MatePad Mini “Reading Edition” features an 8.8-inch flexible OLED display with a 16:10 aspect ratio, a 6400 mAh battery, and a lightweight 255g body measuring just 5.1mm thick. It comes in three colors with a 12GB+256GB configuration, starting at RMB 3,299, available March 28.

    The Huawei MateView GT monitor lineup includes a 27-inch panel with multiple configurations, including standard and Sound editions, as well as smooth and ultra-clear versions. The top-tier model features a 4K soft-light display, 91% screen-to-body ratio, up to 320Hz refresh rate, 95% P3 color gamut, and 100% sRGB coverage. It also supports Huawei’s audio system with a 4.1-channel speaker setup. Prices range from RMB 1,099 to RMB 2,599 depending on configuration.

    The Huawei Vision Smart Screen 6 features a 4.9cm ultra-thin design and a new Huawei Black Crystal display with dual-layer anti-reflective coating, reducing reflectivity to as low as 0.5%. It also uses a sixth-generation LCD panel with 20% improved light transmission. A new “Xiaoyi Sports” feature can recognize players during matches and provide real-time analysis. Prices start at RMB 6,499, with the Vision Smart Screen 6 SE starting at RMB 3,899.

    Under Harmony Intelligent Mobility, Huawei introduced the Shangjie Z7 and Z7T. The vehicles feature a modern design, optional sports packages, ALPS Health Cabin 2.0, Huawei’s dual optical LiDAR system, and an 800V battery platform, offering up to 905 km CLTC range. Pre-orders have begun, starting at RMB 229,800 for the Z7 and RMB 239,800 for the Z7T.

    The Aito M6 has opened pre-orders, with extended-range models starting at RMB 269,800 and fully electric versions at RMB 289,800. It features a 2950mm wheelbase, Huawei Tuling platform, and 700mm wading depth capability, along with standard LiDAR across all models.

    The Luxeed S9T adds a new pink color option, while the S9 introduces a new white color and updated interior designs. Both models support Huawei ADS 4 advanced driving system and LiDAR configurations, starting at RMB 319,800.

    The Luxeed R7 and updated S7 receive new exterior colors, interior upgrades, and enhanced safety features, with optional LiDAR and battery upgrades up to 100kWh. Prices start at RMB 259,800 for the R7 and RMB 239,800 for the S7.

    Additionally, Huawei unveiled its first flagship MPV under Harmony Intelligent Mobility, the Luxeed V9, with showroom units arriving in late April.

    The Aito M7 adds a new LiDAR system, while the M7 Pro+ features an in-cabin vision system, with city-level assisted driving updates rolling out in April. Prices start at RMB 309,800.

    The Aito M8 introduces new colors and upgraded hardware, including a larger battery and extended range of up to 1,590 km combined. Prices start at RMB 369,800.

    DJI sues Insta360 over six patent ownership disputes

    On March 23, according to Jiemian News, DJI filed a lawsuit against Insta360 in the Shenzhen Intermediate People’s Court of Guangdong Province, involving six patent ownership disputes, mainly related to drone flight control, structural design, and imaging processing. In its complaint, DJI stated that the patents in question were inventions created by a former employee within one year after leaving the company, and were closely related to the employee’s job responsibilities while at DJI. Insta360 CEO Liu Jingkang responded on Weibo the same afternoon, stating that the R&D process was compliant and legitimate, and that the case is currently awaiting normal evidence collection and investigation by the court. Liu said that existing evidence shows the patents filed during that period were internally developed ideas and independent innovations within Insta360; concealing inventors in domestic patent filings is a common practice, mainly to avoid talent poaching, and was not specifically done for this former DJI employee; he also noted that some DJI products fall within Insta360’s patent protection scope, but Insta360 will not initiate legal action unless it is blocked from launching new products or innovations in that category. Source

    Douyin releases “Douyin Online Abuse Governance Guidelines (2026 Edition)”

    On March 23, Douyin released the “Douyin Online Abuse Governance Guidelines (2026 Edition),” which for the first time clearly defines the core elements for identifying online abuse, establishes four fundamental governance principles, and outlines circumstances requiring stricter regulation. Under the new guidelines, the platform will assess abusive content and behavior based on three key characteristics: targeting, harm, and scale, while also considering factors such as the type of incident, target subject, number of participants, content of the information, posting frequency, duration, and user reports or complaints. The guidelines also specify four governing principles: the principle of harm prevention, the principle of special protection, the principle of privacy and personal information protection, and the principle of balancing public interest. Among these, the special protection principle emphasizes prioritizing vulnerable groups such as minors, the elderly, and people with disabilities; for public figures and individuals involved in matters of public interest, reasonable public scrutiny should be allowed to prevent overgeneralization of the concept of “online abuse.” The guidelines also clearly define what constitutes “serious circumstances.” Source

    News Worth a Quick Look

    • In a recent email statement to Reuters, an OpenAI spokesperson said that the company will begin showing ads to all ChatGPT Free and Go users in the United States within the next few weeks. Earlier this month, advertising technology company Criteo also stated that OpenAI has integrated it into an ad testing program for ChatGPT Free and Go tiers in the U.S. Criteo primarily provides advertisers with ad purchasing and targeted delivery optimization tools. As ChatGPT usage continues to surge, OpenAI has been exploring advertising as a new revenue stream. Amid intensifying competition in generative AI and rising infrastructure costs, the company is seeking to diversify its income sources. Source
    • According to Cailian Press, XPeng Motors issued an internal announcement on March 23 regarding the establishment of a Robotaxi business unit and related personnel appointments. The unit will operate as a primary organizational division, led by Yuan Tingting, and will include two sub-departments: a product team and a project integration team. XPeng Chairman He Xiaopeng previously stated during the company’s 2025 Q4 and full-year earnings call that XPeng plans to launch pilot Robotaxi passenger services in the second half of this year. Source
    • According to The Wall Street Journal, Meta CEO Mark Zuckerberg is developing a CEO-level AI agent to assist with his work. Sources say the agent enables him to access internal information more quickly, which previously required multiple layers of communication. Source
    • Microsoft has rolled back an update related to virtual accounts that caused an outage in Exchange Online services. While services have been temporarily restored, the root cause is still under investigation, with the issue internally tracked as EX1256020. Source
    • NetEase Cloud Music announced on March 23 that it has integrated OpenClaw, packaging core capabilities such as music recommendations and search into standardized CLI tools and automated skills that users can invoke directly. Source
    • On March 24, Apple announced that its annual Worldwide Developers Conference (WWDC) will be held online from June 9 to June 13 (Beijing time). Source
  • SSPAI Morning Brief: Microsoft Improves Windows 11 as Android Tightens App Sideloading Rules

    SSPAI Morning Brief: Microsoft Improves Windows 11 as Android Tightens App Sideloading Rules

    Morning Brief

    1. Microsoft promises multiple improvements to Windows 11
    2. Android may require a “cooling-off” day before installing apps from outside official stores
    3. Halide CEO accuses former co-founder poached by Apple
    4. U.S. tech companies see rising competition in “token farming”
    5. Meta’s metaverse plans move toward an end
    6. Preprint platform arXiv to operate independently from Cornell University
    7. News Worth a Quick Look

    Microsoft promises multiple improvements to Windows 11

    On March 20, Pavan Davuluri, Executive Vice President of Microsoft’s Windows and Devices division, announced that Windows 11 will receive major quality upgrades. The promised measures include—

    • The taskbar will once again support docking at the top, left, or right of the screen;
    • AI integration will be more restrained, with unnecessary Copilot shortcuts removed from core apps such as Screenshot, Photos, Widgets, and Notepad;
    • Users will be able to skip updates during initial setup on new devices, and in daily use can choose to shut down or restart without installing updates;
    • Core Windows experience components, including the Start menu, will be migrated to the WinUI3 framework to reduce interaction latency, with a focus on improving File Explorer launch and search performance;
    • The Widgets panel will adopt a quieter, less intrusive default design;
    • Stability improvements will be made to Windows Subsystem for Linux (WSL), Windows Hello biometric authentication, and peripheral connectivity.

    As background, Windows 11 has recently pushed AI features heavily across its interface while also facing frequent stability issues, leading to user complaints. These optimizations will be rolled out for testing to Windows Insider Program members between March and April this year.

    Android may require a “cooling-off” day before installing apps from outside official stores

    On March 19, Google unveiled a new proposal targeting Android app “sideloading” (installing apps from non-official stores), which includes a so-called mandatory 24-hour cooling-off period.

    According to the developer identity verification policy announced by Google last August, developers will be required to submit real names and identity verification documents to the platform. This mandatory verification will first be implemented in countries such as Brazil and Singapore starting this September, with a global rollout planned for 2027. The new sideloading process, including the cooling-off period, will be made available earlier in August as an optional setting for advanced users. The high threshold has drawn widespread criticism, and Google later promised in November to continue allowing sideloading for power users.

    Under the newly outlined process, users who wish to enable sideloading must first activate Developer Mode, confirm they are not being coerced, reboot their device, and then wait 24 hours before verifying their identity via fingerprint or facial recognition. Once completed, users can choose to enable sideloading for either 7 days or indefinitely. Google states that the reboot step is intended to interrupt potential scammer activities such as ongoing calls or remote screen monitoring, while the cooling-off period is designed to break the sense of urgency often exploited by fraudsters, giving users time to think more carefully.

    Halide CEO accuses former co-founder poached by Apple

    According to The Information, after failing to acquire the developer behind the well-known photography app Halide, Apple instead recruited one of the startup’s co-founders. The move has since triggered a fierce legal dispute within the original team, with both sides accusing each other of misappropriating funds and stealing trade secrets.

    The lawsuit was filed by Ben Sandofsky, CEO of Halide’s parent company Lux Optics, in the Superior Court of Santa Cruz County, California. He alleges that former co-founder and designer Sebastian de With misappropriated over $150,000 of company funds for personal expenses since 2022, and, after joining Apple, disclosed confidential materials and source code to the company. De With was dismissed in December last year following these allegations and officially joined Apple’s design team in January this year. In response, de With’s legal counsel strongly denied any intellectual property theft and counter-accused Sandofsky of financial misconduct, claiming the lawsuit is retaliation for de With’s request to review company financial records.

    The dispute stems from acquisition talks between Apple and Lux Optics last summer. Apple had initially hoped to strengthen its native iOS camera app through the acquisition, explicitly stating that Lux’s intellectual property was a key consideration. However, the Lux team believed that future app updates could secure a better valuation, leading to the collapse of negotiations. Just two months later, Apple began recruiting de With independently. Apple is not named as a defendant in the case.

    Sources indicate that upgrading the native camera app is currently a top priority within Apple. To enable the upcoming iPhone 18 Pro to rival professional-grade cameras in certain advanced features, Apple is actively seeking to absorb both the underlying technical expertise and design talent from specialized imaging apps.

    U.S. tech companies see rising competition in “token farming”

    According to The New York Times, major Silicon Valley companies such as Meta, Shopify, and OpenAI have recently begun incorporating AI usage into employee performance evaluations, sparking an internal competition among programmers dubbed “token farming” (or “Tokenmaxxing”). In an effort to demonstrate productivity to management, some employees are aggressively invoking AI tools, with individual usage reportedly exceeding $150,000 worth of compute in a single month. One OpenAI engineer processed as many as 210 billion tokens in just one week—equivalent to roughly 33 full Wikipedia texts.

    Traditional prompt-and-response usage of AI is unlikely to reach such scale. Instead, employees are primarily relying on the rising class of agentic programming tools. These systems can operate autonomously around the clock without human supervision, spawning multiple sub-agents to handle tasks in parallel—continuing to consume vast amounts of compute even while the user sleeps.

    According to reports, generous token budgets have effectively become a new form of employee perk, akin to free lunches. To encourage adoption of new technologies, companies have introduced internal leaderboards tracking AI consumption and even reward top users. This incentive structure has also given rise to opportunistic behavior.

    Industry observers note that behind this frenzy of compute consumption lies a growing sense of “token anxiety” among white-collar workers. As AI becomes increasingly capable of independently writing code, programmers face the looming threat of replacement. To avoid falling behind in the wave of automation, employees are treating the operation of large-scale AI agent systems as a form of career self-defense—signaling to management that they are keeping pace with cutting-edge technology, even as the actual quality of code output takes a back seat.

    Meta’s metaverse plans move toward an end

    According to The New York Times, Meta has effectively brought an end to Mark Zuckerberg’s virtual reality (VR)-based metaverse vision. Last week, Meta announced that starting June 15, users will no longer be able to access its “metaverse social app” Horizon Worlds via VR headsets. Although Meta later clarified that it will continue to support some existing VR experiences within the app, it made clear that no new applications will be added.

    Behind this retreat is a comprehensive shift of resources toward AI. The division responsible for metaverse projects has laid off around 10% of its staff in recent months. At the latest developer conference, Zuckerberg mentioned the “metaverse” only twice, compared to 23 mentions of “AI.” He is now betting on “superintelligence,” with Meta expected to invest at least $115 billion this year, primarily in AI development and large-scale data center infrastructure. Meanwhile, the company’s hardware focus has shifted to AR smart glasses equipped with built-in AI assistants.

    Meta acquired VR headset company Oculus in 2014 for $2 billion, rebranded itself as Meta in 2021, and has accumulated approximately $80 billion in losses on its metaverse initiative. During the pandemic, the demand for remote work and contactless social interaction briefly propelled the metaverse concept into the spotlight. However, Meta’s virtual world repeatedly ran into technical limitations, with its crude floating avatars becoming a frequent subject of online ridicule. Additionally, the high cost of hardware has hindered broader adoption. Even Apple’s Vision Pro has yet to break into the mass consumer market, and VR has not replaced smartphones as the next computing platform.

    Preprint platform arXiv to operate independently from Cornell University

    According to Science, the globally renowned preprint platform arXiv has announced that it will officially separate from Cornell University on July 1 and become an independent nonprofit organization. The move aims to broaden funding channels in response to the surge in paper submissions in recent years and to invest resources in combating “AI slop.” arXiv allows scientists to share research findings quickly and freely before formal peer review, serving as a foundational platform for scholarly communication in fields such as physics, computer science, and mathematics.

    Since its founding in 1991, the platform has grown rapidly, with the recent AI boom significantly accelerating submission volumes—expected to exceed 300,000 papers this year. To cope, the arXiv team has expanded to 27 members, driving its annual operating costs to $6.7 million and resulting in deficits over the past two years. Cornell officials stated that spinning off into an independent entity would alleviate concerns among potential donors about funds being controlled by the university, enabling arXiv to more directly attract financial support from libraries, research institutions, and foundations worldwide.

    In response to the transition, some researchers have expressed concerns on social media, including criticism of the proposed CEO salary of up to $300,000 and worries about whether the platform might eventually be acquired by a commercial entity. Management responded that arXiv will maintain its successful nonprofit model, with no noticeable changes for users in the short term, and no immediate increase in institutional membership fees. Cornell University, the Simons Foundation, and others have already secured initial funding to ensure the new organization begins operations without a deficit.

    Last year, bioRxiv and medRxiv—preprint platforms in biology and medicine inspired by arXiv—also separated from their academic hosts for similar reasons. As arXiv founder Paul Ginsparg has noted, while universities offer many advantages, historical experience suggests they are not well-suited to provide long-term, comprehensive support for global research infrastructure of this kind.

    News Worth a Quick Look

    • According to PBX SCIENCE, Ubuntu 26.04 LTS, set to be released on April 23, will break a 46-year Linux tradition: when users enter their password with the sudo command, the terminal will display asterisks (*) to indicate input, instead of remaining blank. Since its introduction in 1980, sudo has provided no visual feedback during password entry, originally to prevent observers from inferring password length. In this update, Ubuntu adopts sudo-rs, a Rust-based reimplementation of sudo focused on security. The development team argues that modern graphical login interfaces already display password placeholders, making the terminal’s invisible input offer minimal security benefits, while often confusing newcomers into thinking the terminal has frozen or their keyboard is unresponsive.
    • According to The Wall Street Journal, OpenAI plans to integrate ChatGPT, Codex, and the Atlas browser into a single desktop “super app.” The move aims to streamline user experience, consolidate internal engineering resources, and refocus strategy on enterprise clients and developers. In an internal memo, OpenAI’s Head of Applications, Fidji Simo, noted that the team’s efforts are currently too fragmented, slowing development and making quality harder to maintain. The new app will emphasize AI agent capabilities. The existing ChatGPT mobile app will remain unchanged.
    • According to The Verge, Google is testing AI-generated rewrites of webpage titles in search results. The outlet found that titles of multiple articles were replaced with AI-generated versions without any indication of modification. A Google spokesperson confirmed this is a “limited” experiment using generative AI, intended to better match user intent and improve engagement. However, Google stated that if the feature is officially launched, it will not rely on generative models—without explaining how titles would then be rewritten. Google has previously adopted a similar approach in its Discovery feed.
    • According to Reuters, more than a decade after the failure of the Fire Phone, Amazon is planning a return to the smartphone market under the codename “Transformer.” The company aims to create a mobile device deeply integrated with AI and the Alexa voice assistant, focusing on AI-native features. Amazon is exploring multiple design directions, including minimalist “dumbphone”-style devices inspired by the Light Phone.
    • On March 21, the co-founder and COO of the AI coding app Cursor confirmed that its new model, Composer-2, is fine-tuned from Moonshot AI’s Kimi K2.5 model. Cursor released Composer-2 on March 19, claiming performance between OpenAI GPT-5.4 (high) and Anthropic Claude Opus 4.6 (high) at a lower cost. However, developers noticed the model ID in code as “Kimi-k2p5,” leading to speculation that Composer-2 was essentially a reinforced-learning version of K2.5. Cursor acknowledged that failing to mention Kimi as the base model in its launch blog was an oversight and said it would correct this in future releases. Kimi responded that Cursor accesses K2.5 via the FireworksAI inference and deployment platform under a licensed commercial partnership, and welcomed its integration into the broader open model ecosystem.
    • On March 21, China’s Cyberspace Administration stated that current standards for labeling short video content across platforms are inconsistent. Some videos involving fictional narratives, staged marketing, or AI-generated content lack proper labeling, misleading the public, disrupting social order, and polluting the online ecosystem. In response, regulators plan to standardize content labeling: first, by defining required label categories; second, by making labeling a mandatory step before publishing; and third, by retroactively reviewing and labeling existing content in phases. Authorities will soon issue detailed requirements and timelines, implement nationwide enforcement, and conduct inspections—penalizing accounts and platforms that fail to comply, with public disclosure of violations.
    • On March 22, Tencent announced the launch of a new tool called ClawBot, officially integrating OpenClaw into WeChat. Once added, ClawBot will appear in users’ contact lists like a regular contact. Earlier this month, Tencent also introduced QClaw for individual users, Lighthouse for developers, and WorkBuddy for enterprise services.
  • SSPAI Morning Brief: Xiaomi Launches SU7 EV as MiniMax Unveils M2.7 AI Model

    SSPAI Morning Brief: Xiaomi Launches SU7 EV as MiniMax Unveils M2.7 AI Model

    Morning Brief

    1. Xiaomi launches multiple new products
    2. MiniMax releases MiniMax-M2.7 model
    3. Google Stitch introduces “vibe design” concept
    4. Philips unveils new Moving Sound series
    5. Sina Cloud announces permanent shutdown on September 16
    6. ASUS launches ExpertCenter PN55 mini PC
    7. LG introduces two ultrawide monitors
    8. News Worth a Quick Look

    Xiaomi launches multiple new products

    On March 19, Xiaomi unveiled the new-generation SU7, the latest Xiaomi Laptop Pro 14, and the Xiaomi Watch S5.

    The new-generation SU7, as the first major facelift of the model, brings comprehensive upgrades in design, performance, intelligence, and safety. It introduces new color options and premium configurations, and is equipped with an upgraded V6s Plus motor and a silicon carbide high-voltage platform. The Max version accelerates from 0 to 100 km/h in 3.08 seconds, reaches a top speed of 265 km/h, and offers a maximum CLTC range of 902 km. The vehicle also comes standard with LiDAR, NVIDIA Thor chip, and Xiaomi HAD advanced driver assistance system, supporting large-model-driven intelligent driving capabilities. Pricing across the lineup has increased by RMB 4,000 compared to the previous generation, with a starting price of RMB 219,900.

    The Xiaomi Laptop Pro 14, released alongside, focuses on a balance of portability and high performance. It weighs approximately 1.08 kg, can be configured with up to an Intel Core Ultra X7 358H processor, features a 3.1K OLED touchscreen, and offers up to around 19.8 hours of battery life, starting at RMB 8,499. The Xiaomi Watch S5 is positioned as a long-battery-life sports smartwatch, featuring a stainless steel body and AMOLED display, supporting dual-frequency GNSS, health tracking, and up to 21 days of battery life. It also integrates deeply with Xiaomi vehicles, with a starting price of RMB 1,199. Source

    MiniMax releases MiniMax-M2.7 model

    On March 18, MiniMax announced the launch of its flagship agent model, MiniMax-M2.7. The model focuses on enabling recursive self-iteration. According to MiniMax, it “builds an autonomous evolving Agent Harness system that allows the model to deeply participate in its own training, optimization, and reinforcement learning processes.”

    Test results show that the model performs strongly in engineering development and professional productivity tasks. For example, it achieved 56.22% on the SWE-Pro benchmark, 55.6% on VIBE-Pro, and 57.0% on Terminal Bench 2.

    In terms of specifications, MiniMax-M2.7 supports up to a 204.8K context window, making it suitable for handling large codebases and long-chain conversational tasks. The standard version delivers output speeds of 60 tokens per second, while the highspeed version increases output to 100 tokens per second without compromising performance.

    In pricing, MiniMax-M2.7 costs RMB 2.1 per million input tokens, RMB 8.4 per million output tokens, RMB 0.42 per million cached reads, and RMB 2.625 per million cached writes. The MiniMax-M2.7-highspeed version is priced at RMB 4.2 per million input tokens, RMB 16.8 per million output tokens, with the same caching costs as the standard version. Source

    Google Stitch introduces “vibe design” concept

    On March 18, Google announced an update to its UI design tool Stitch, introducing a new design philosophy called “vibe design.” The update includes a redesigned user interface and adds a native AI-powered infinite canvas that supports multiple context types such as images, text, and code, providing greater flexibility from early ideation to final prototyping.

    AI can not only generate new pages by “interviewing” users but also offer real-time design evaluations and adjust outputs dynamically based on prompts such as “give me three menu options” or “show this screen in different color palettes.” Google stated that with the addition of the Design Agent and Agent Manager, the system can analyze the entire project lifecycle, perform logical reasoning, and track multiple parallel creative workflows. Users can also assign specific tasks directly on the canvas to collaboratively tackle more complex design challenges.

    Additionally, Stitch now supports one-click extraction of design systems from any URL. To improve cross-platform compatibility, it introduces a DESIGN.md format for agents, enabling users to synchronize design rules and commonly used tools, and to import, export, or reuse designs across different Stitch projects.

    Google also emphasized that developers can now integrate external AI tools such as Cursor, Antigravity, or Gemini CLI into their projects via the Stitch MCP server and SDK, further expanding Stitch’s application scenarios. Source

    Philips unveils new Moving Sound series

    On March 18, Philips officially introduced the new Moving Sound series. The lineup adopts high-saturation color schemes popular in the 1980s, giving it a distinctly retro design style.

    The series includes two wireless speakers, both supporting Bluetooth 5.5 and Auracast, featuring IP67 protection and up to approximately 24 hours of battery life. They are equipped with neon-colored buttons and multi-color LED lighting effects. Both models feature color displays that simulate cassette tape animations while showing playback information. The Tube model delivers 140W output, equipped with dual woofers, dual tweeters, and passive radiators, emphasizing bass performance. The Roller model offers 60W output, with a smaller form factor focused on portability.

    The lineup also includes two headphone products. The Ringo Duo is a lightweight over-ear headphone inspired by classic Walkman-style designs, featuring a metal headband and supporting both Bluetooth and USB-C connections, with up to 26 hours of battery life. The Buds true wireless earbuds feature a six-microphone system for active noise cancellation and call pickup. With ANC turned off, they offer up to 42 hours of battery life with the charging case. The case is relatively large and includes a display capable of showing animations. Source

    Sina Cloud announces permanent shutdown on September 16

    On March 16, Sina Cloud announced on its official website that, due to business adjustments, the platform will be shut down at 24:00 on September 16, 2026, with all services permanently terminated. According to the announcement, from now until September 16, 2026, Sina Cloud will stop new user registrations, new service activations, and cloud credit top-ups. During this period, any remaining cloud credits and points in user accounts can still be used normally.

    For users with active services, the company advises completing service migration, data export, backup, and account cleanup before September 16, 2026. After that date, all platform services will cease, and users must make the necessary preparations in advance. Source

    ASUS launches ExpertCenter PN55 mini PC

    On March 18, ASUS officially introduced the ExpertCenter PN55 mini PC. The device is powered by AMD’s Gorgon Point processors, offering multiple configurations ranging from consumer-grade Ryzen AI 5 430 to Ryzen AI 9 HX 470, as well as business-oriented Ryzen AI 7 PRO 450 and Ryzen AI 5 PRO 440 options.

    The PN55 measures 130 × 130 × 34 mm and weighs just 530 grams. Internally, it provides two DDR5-5600 SO-DIMM memory slots, two M.2 2280 PCIe Gen4 ×4 NVMe SSD slots, and includes a wireless network card.

    In terms of external I/O, the front panel includes two USB-A ports, one USB-C port (all supporting 10 Gbps), and a 3.5 mm combo audio jack. The rear panel features one full-function USB-C (40 Gbps), one USB-A (10 Gbps), one USB-A (480 Mbps), one HDMI 2.1 FRL, two DisplayPort 1.4 ports, and two 2.5 GbE RJ45 Ethernet ports.

    Additionally, the side of the device includes a Kensington lock slot, while the top integrates a Windows Hello fingerprint sensor. Source

    LG introduces two ultrawide monitors

    On March 18, LG introduced two new monitors: the 34U640B-B and 34U620B-B. Both models feature 34-inch 1800R curved VA panels with a UWQHD resolution of 3440 × 1440 and a 144 Hz refresh rate.

    Both displays offer a 5 ms response time, 300 nits brightness, a 4000:1 contrast ratio, and cover 99% of the sRGB color gamut. They also support HDR10, variable refresh rate (VRR), and adaptive sync technology. Additionally, both monitors include low blue light and flicker-free eye-care features, with stands that support height adjustment and tilt.

    The main difference lies in connectivity. In addition to the two HDMI ports and one DisplayPort found on the 34U620B-B, the 34U640B-B also includes a full-function USB-C port with 65W Power Delivery and two USB-A 5 Gbps ports. Source

    News Worth a Quick Look

    • Brazil’s antitrust regulator CADE has intensified its investigation into Apple’s restrictions on iPhone NFC functionality and has issued a formal notice requiring the company to provide details on its fee structure, technical requirements, and developer agreements. Apple has been asked to respond by March 30. Previously, Brazil’s central bank and banking associations filed complaints alleging that Apple’s limitations on third-party access to iPhone NFC may constitute unfair competition. Apple has argued that its market share in Brazil’s smartphone market is only around 10%, and stated that it has opened NFC access to third parties since 2024, maintaining that its service fees do not violate local laws. The company also emphasized that Brazil’s payments market remains highly competitive and that Apple Pay does not exclude competitors. However, regulators have not accepted this explanation. The investigation also involves Brazil’s mainstream payment system PIX, with Apple’s refusal to support its contactless payment protocol becoming a key point of controversy. Source
  • SSPAI Morning Brief: Midjourney V8 Launches as Xiaomi Unveils MiMo-V2-Pro AI Model

    SSPAI Morning Brief: Midjourney V8 Launches as Xiaomi Unveils MiMo-V2-Pro AI Model

    Morning Brief

    • Tencent launches QClaw
    • XPeng P7 introduces Max version and three new color options
    • Xiaomi releases MiMo-V2-Pro model
    • Midjourney unveils V8 model
    • Firefox to introduce mascot “Kit”
    • FIFA partners with YouTube for World Cup content
    • Horizon Worlds to discontinue VR version

    Tencent launches QClaw

    On March 18, Tencent announced the public beta launch of QClaw, alongside a new version update that upgrades the WeChat entry into a mini program, enabling users to upload or receive files from their computers. A new “Inspiration Square” has also been introduced, featuring preset common tasks and skills, allowing users to execute actions with one click without writing commands.

    QClaw is a local AI agent one-click launch package developed by Tencent PC Manager team, based on the OpenClaw framework. Its core highlight is the integration of both WeChat and QQ entry points, allowing users to remotely control their local computers by simply sending natural language commands within chat windows—without needing to open a dedicated client. It supports tasks such as file processing, data analysis, and browser automation.

    The platform supports both Windows and macOS, integrates mainstream Chinese large models including Kimi, MiniMax, GLM, and DeepSeek, and connects to the ClawHub skill ecosystem, offering more than 5,000 predefined skills. Source

    XPeng P7 introduces Max version and three new color options

    On March 18, XPeng Motors held a product launch event, announcing three new exterior color options for the P7 series—Stellar Red, Midnight Black, and Nebula White—bringing the total number of available colors to ten. At the same time, the P7 Max version was introduced, offering three configurations: 702 Max, 820 Max, and 750 AWD Max, with prices ranging from RMB 203,800 to RMB 243,800.

    The new model offers an intelligent upgrade package, allowing users to choose different configurations of Turing AI chips as well as second-generation VLA and VLM models. Two new interior color options—Gravity Red and Lava Gray—have also been added, along with “Stellar Knight” and “Dark Knight” styling packages. Source

    Key product specifications chart; image courtesy of the news source

    Xiaomi releases MiMo-V2-Pro model

    On March 19, Xiaomi unveiled its flagship foundation model for agent scenarios, Xiaomi MiMo-V2-Pro. According to the company, the model has over 1 trillion total parameters, with 42 billion activated parameters, and adopts a hybrid attention architecture. It supports up to 1 million tokens of context and is designed for high-intensity agent workflows, long-horizon planning, and complex tool usage in real-world scenarios.

    Xiaomi stated that MiMo-V2-Pro ranks eighth globally and second in China on the Artificial Analysis leaderboard. It has demonstrated end-to-end task execution capabilities in agent frameworks such as OpenClaw and Claude Code, capable of completing complex workflows and precise tool calls without human intervention. The company claims its overall performance surpasses Claude Sonnet 4.6 and approaches Opus 4.6, while its API pricing is only about one-fifth of the latter.

    In terms of capabilities, MiMo-V2-Pro continues the Hybrid Attention mechanism of its predecessor, increasing the ratio from 5:1 to 7:1, scaling up parameters while maintaining inference efficiency. Through expanded compute during post-training across broader agent tasks, the model extends beyond dialogue and code generation into full task execution. Xiaomi also revealed that the anonymous model previously released on OpenRouter under the name Hunter Alpha was an early internal test version of MiMo-V2-Pro.

    MiMo-V2-Pro is now available via API, priced at $1 per million input tokens and $3 per million output tokens for contexts up to 256K, and $2 input / $6 output for contexts up to 1M tokens. The model has also been integrated into MiMo Studio, WPS Lingxi, Xiaomi Browser, and the mobile agent product miclaw. Source

    Midjourney unveils V8 model

    On March 18, the Midjourney team opened early access to the V8 model in its user community, inviting users to test the model and provide feedback. According to Midjourney, V8 shows significant improvements in understanding complex prompts, delivers better image detail and overall consistency, and further enhances text rendering capabilities. Generation speed has also increased by approximately five times compared to previous versions.

    Alongside the V8 release, Midjourney has upgraded its web interface, introducing a more natural conversational creation workflow, a new Grid Mode for easier viewing of image sets, and a redesigned sidebar for parameter controls to reduce visual clutter. In terms of features, V8 supports multiple aspect ratios at launch, as well as parameters such as –chaos, –weird, –exp, and –raw, while remaining compatible with V7 personalization settings, mood boards, and style references.

    The new version also introduces a native 2K output mode (–hd) and a –q 4 mode to improve visual consistency. Midjourney noted that Relax mode is not yet supported and is currently awaiting new server clusters. At present, tasks using –hd, –q 4, style references, and mood boards render at approximately one-quarter the speed of standard tasks and are billed at four times the cost.

    Midjourney emphasized that V8 is a brand-new model with its own strengths and limitations, and may require different prompting techniques. Users are encouraged to use longer and more specific prompts, combined with personalization, style references, and mood boards to fully leverage V8’s capabilities, while also using the web-based rating system to help improve the model. Source

    Firefox to introduce mascot “Kit”

    On March 17, Mozilla announced that Firefox will retire its iconic “fox wrapped around the globe” logo, which has been associated with the browser for over 20 years, and introduce a new mascot named Kit.

    Illustration of Kit, image from source

    In its blog post, Mozilla stated that the internet has entered a new era, and users need a new guide that conveys warmth and familiarity. Kit is designed to fill this role as a unified global identity. Conceptually, Kit is defined as a pure representation of Firefox—it is neither the traditional fox nor a Himalayan red panda. The character was created through a collaboration between Mozilla, creative agency JKR, and illustrator Marco Palmieri.

    Kit will appear in Firefox’s onboarding experience, feature announcement pop-ups, and settings confirmation screens, and will also be used across Firefox’s website, blog, and social media platforms.

    Mozilla also clarified that Kit was entirely handcrafted by human designers through hundreds of detailed refinements—such as tail movement, textures, and gradients—and is not generated by AI. Users are advised not to mistake it for a chatbot or AI assistant. Source

    FIFA partners with YouTube for World Cup content

    On March 18, YouTube announced a partnership with the Fédération Internationale de Football Association (FIFA), under which YouTube will serve as an official content partner platform for the 2026 World Cup. During the tournament, YouTube will provide match highlights, archival footage, and a dedicated creator program for global content creators.

    According to current arrangements, rights-holding broadcasters will be allowed to livestream the first 10 minutes of each match on their YouTube channels, and may also publish selected full matches, extended highlights, behind-the-scenes footage, short-form videos, and on-demand content. FIFA will also open access to portions of its historical World Cup archives, including classic matches and iconic moments.

    The collaboration will also include a global YouTube creator program, granting selected creators access to exclusive interviews and behind-the-scenes content during the tournament. The 2026 World Cup is scheduled to take place from June 11 to July 19 across the United States, Canada, and Mexico, featuring an expanded format of 48 teams and a total of 104 matches. Source

    Horizon Worlds to discontinue VR version

    On March 18, Meta confirmed that its metaverse platform Horizon Worlds will gradually phase out support for VR devices, with the VR version set to be discontinued and removed from the Quest Store on June 15.

    According to Meta’s current timeline, starting March 31, the Quest Store will no longer display user-created Horizon Worlds environments and events, and users will be unable to access multiple in-world locations—including Horizon Central, Events Arena, Kaiju, and Bobber Bay—via VR headsets. By June 15, the Horizon Worlds app will be completely removed from Quest devices, after which users will only be able to access the platform in 2D mode via the Meta Horizon app on iOS or Android.

    At the same time, the recently introduced Hyperscape Capture testing feature will also be discontinued. Meta stated that while users will still be able to capture and view Hyperscape content in the future, social features such as sharing, invitations, and multiplayer experiences will no longer be supported. Source

  • SSPAI Morning Brief: OpenAI Launches GPT-5.4 Models as Spotify Adds Lossless Audio and Exclusive Mode

    SSPAI Morning Brief: OpenAI Launches GPT-5.4 Models as Spotify Adds Lossless Audio and Exclusive Mode

    Morning Brief

    1. OPPO unveils Find N6
    2. GAC and Huawei jointly launch “Qijing” GT7
    3. Beats and Nike release co-branded Powerbeats Pro 2
    4. OpenAI releases GPT-5.4 mini and nano models
    5. Nintendo Switch rolls out handheld enhancement mode update
    6. Nikon announces recall of certain Z5II full-frame mirrorless cameras
    7. Spotify Windows version adds exclusive mode and lossless audio support
    8. Kagi launches Small Web mobile app and browser extension
    9. News Worth a Quick Look

    OPPO unveils Find N6

    On March 17, OPPO introduced its foldable flagship, the OPPO Find N6. The device is powered by the Snapdragon 8 Gen 5 Elite mobile platform and features an 8.12-inch inner display alongside an outer display with ultra-narrow bezels. Both screens support 1-nit eye-care brightness and a global peak brightness of 1800 nits. It adopts a titanium alloy “Sky Dome” hinge and “Sky Dome” memory glass, passing TÜV Rheinland’s 600,000-fold durability test, delivering a virtually crease-free experience both out of the box and after prolonged use. It is equipped with a 6000mAh “Glacier” battery, supporting 80W wired and 50W wireless fast charging. For imaging, it features a Hasselblad-tuned 200MP quad-camera system, including a 70mm portrait periscope telephoto lens and a Danxia color reproduction lens. The device also supports IP56/IP58/IP59 water resistance and includes a “Shanhai” communication enhancement chip with support for 66 frequency bands.

    The OPPO Find N6 is available in Gold Orange, Raw Titanium, and Deep Black. Pricing starts at RMB 9,999 for the 12GB + 256GB version, RMB 10,999 for 16GB + 512GB, and RMB 11,999 for 16GB + 1TB. It will officially go on sale globally starting March 20. Source

    GAC and Huawei jointly launch “Qijing” GT7

    On March 17, Huawei Qiankun and GAC Group unveiled a jointly developed high-end intelligent vehicle brand, AISTALAND, in Guangzhou, and introduced its first model, the AISTALAND GT7. The GT7 measures 5050 × 1980 × 1470 mm with a 3000 mm wheelbase, featuring a 21.5° “shooting brake angle” and a 51.2% wheel-to-body ratio. It comes equipped with 275 mm wide tires, frameless side mirrors, and semi-hidden door handles. In terms of hardware, it features Huawei’s 896-line dual-path image-grade LiDAR, a triple-motor system, and CATL’s 800V/6C Qilin battery. It also integrates Huawei Qiankun Intelligent Driving ADS, the HUAWEI XMC digital chassis engine, an 88-inch HUAWEI XHUD, and a triple-layer silver PDLC smart dimming panoramic roof.

    The AISTALAND GT7 is now open for pre-orders with a refundable deposit of RMB 999, with official launch and deliveries planned within 2026. Source

    Beats and Nike release co-branded Powerbeats Pro 2

    On March 17, Beats and Nike jointly announced their first co-branded hardware product, the Nike Special Edition Powerbeats Pro 2. This version features a fluorescent yellow colorway and marks the first time both brands share visible logos, with the Nike Swoosh on the right earbud and the Beats logo on the left. It also comes with a matte black charging case accented with fluorescent yellow speckles, with a Nike slogan engraved inside the lid.

    Powerbeats Pro 2 supports adaptive active noise cancellation and transparency mode, and includes built-in heart rate monitoring for workouts, which can sync with fitness apps. It offers IPX4 sweat and water resistance and features a secure ear-hook design, with up to 45 hours of total battery life including the charging case. The Nike Special Edition Powerbeats Pro 2 will officially go on sale on Friday, March 20, priced at RMB 2,099, available via Apple.com and the Apple Store at Shanghai iapm. Source

    OpenAI releases GPT-5.4 mini and nano models

    On March 17, OpenAI announced two smaller models in the GPT-5.4 series: GPT-5.4 mini and GPT-5.4 nano. GPT-5.4 mini delivers more than 2× the speed of GPT-5 mini and demonstrates performance close to the full GPT-5.4 model in benchmarks such as SWE-Bench Pro and OSWorld-Verified, with particular improvements in coding, multimodal understanding, and computer-use capabilities. GPT-5.4 nano, as the smallest and most cost-effective model in the lineup, is designed for tasks such as data classification, extraction, and simple sub-agent workflows.

    Both models offer a 400K context window. GPT-5.4 mini is priced at $0.75 per million input tokens and $4.50 per million output tokens via API. GPT-5.4 nano is available only via API, priced at $0.20 per million input tokens and $1.25 per million output tokens. GPT-5.4 mini is now available across the API, Codex, and ChatGPT (for Free and Go users). Source

    Nintendo Switch rolls out handheld enhancement mode update

    On March 17, Nintendo released system firmware version 22.0.0 for Switch and Switch 2 consoles. The update introduces a “Handheld Mode Enhancement” feature, which can be enabled under “Settings – System – Software Operation Mode.” Once enabled, software in handheld mode can run with performance equivalent to TV mode. The update also refines social and management features, adding private friend notes that can be edited on both the console and mobile app (version 3.3.0 or later), and allowing invitations to friends who have not completed initial setup to join game chats. Parental controls have been upgraded as well—when entering a PIN on the console, linked apps (version 2.4.0 or later) will receive real-time notifications. In terms of interface details, the HOME menu’s virtual game card loading animation has been redesigned; full-screen video playback in eShop and game news now supports 10-second skip forward/back using ZL/ZR buttons; edited album content is now included in automatic uploads; and airplane mode can retain Bluetooth, Wi-Fi, and NFC settings. Accessibility features have also been expanded with multi-language text-to-speech support, though Chinese is not yet supported. Source

    Nikon announces recall of certain Z5II full-frame mirrorless cameras

    On March 17, Nikon announced a recall of certain Z5II full-frame mirrorless cameras. The recall is due to specific components in a small number of units that may, in rare cases, affect normal operation. Nikon has identified the affected serial number ranges, and users can check eligibility on the official website by entering their camera’s serial number. Nikon will provide free repairs for affected units and cover round-trip shipping costs, regardless of warranty status. Users can contact Nikon customer support or authorized service centers before sending in their devices. Additionally, some Z6III and ZR cameras purchased outside mainland China are also included in the recall. Source

    Spotify Windows version adds exclusive mode and lossless audio support

    On March 17, Spotify announced the rollout of “Exclusive Mode” for Premium users on its Windows desktop client. Combined with the newly introduced Lossless audio tier, the feature enables “bit-perfect” output at up to 24-bit/44.1 kHz FLAC quality. When enabled, Spotify takes full control of the system’s audio processing, bypassing the Windows mixer’s resampling and volume adjustments, and delivers the original audio stream directly to an external DAC or audio interface. While in Exclusive Mode, other applications cannot output sound through the same device. The feature does not support video, podcasts, or Canvas clips, and automatic mixing and crossfade are disabled when activated. Users can enable the feature under “Settings – Playback,” selecting a compatible output device and turning on “Normalize Volume” to maintain maximum output. The feature is currently available only on Windows, with a macOS version expected in the future. Source

    Kagi launches Small Web mobile app and browser extension

    On March 17, search engine Kagi announced the release of the Small Web mobile app and browser extension for iOS and Android. Small Web aims to bring human-created, non-commercial content—such as personal blogs, web comics, and independent videos—to mobile platforms. The initiative originated from Kagi’s 2023 project of the same name and now indexes over 30,000 such sites. The new app and extension support filtering results by categories such as videos, code repositories, and comics, and offer features like random discovery, distraction-free reading mode, and bookmarking.

    Although some users on Hacker News have pointed out that its filtering criteria—such as requiring an actively updated RSS feed—may exclude certain single-purpose or experimental websites, Kagi maintains that the project is intended to counter the growing impact of AI-generated content on the web ecosystem. Users can currently submit suggestions for human-created sites via its GitHub page. Source

    News Worth a Quick Look

    • On March 16, Dell announced the reinstatement of its Pro Precision workstation product line, reversing its previous rebranding under the Dell Pro Max lineup. The series includes both desktop and mobile models. The flagship tower workstation, Pro Precision 9 T6, features 15 PCIe slots and supports up to two 600W or five 300W NVIDIA RTX PRO Blackwell professional GPUs. The Pro Precision mobile workstations are available in 14-inch and 16-inch configurations. Source
    • On March 17, Apple officially released its first “Background Security Update,” delivering lightweight security patches for Safari, the WebKit framework stack, and other system libraries. The update supports iOS 26.1, iPadOS 26.1, and macOS 26.1, and is designed to address WebKit-related vulnerabilities. Background Security Updates can be downloaded automatically, but require a device restart to complete installation, which takes about one minute. Users can view update details under “Settings – Privacy & Security.” Source
  • SSPAI Morning Brief: NVIDIA Unveils DLSS 5 as Apple Launches AirPods Max 2

    SSPAI Morning Brief: NVIDIA Unveils DLSS 5 as Apple Launches AirPods Max 2

    Morning Brief

    1. Apple unveils AirPods Max 2
    2. NVIDIA introduces DLSS 5
    3. Apple Watch in mainland China rolls out mobile atrial fibrillation (AFib) irregular rhythm notification feature
    4. Zhipu releases GLM-5-Turbo
    5. National Satellite Internet System and Service Standardization Technical Committee approved for establishment
    6. China Academy of Information and Communications Technology releases Blue Book on Next-Generation Smart Terminals (2025)
    7. News Worth a Quick Look

    Apple unveils AirPods Max 2

    On March 16, Apple introduced AirPods Max 2. The new model is powered by the H2 chip, delivering stronger active noise cancellation with a 1.5× improvement over the previous generation, more effectively reducing noises such as airplane engines or train movement. In addition, the H2 chip brings new features to AirPods Max 2, including Adaptive Audio, Conversation Awareness, Voice Isolation, camera remote control, studio-quality audio recording, loud sound reduction, personalized volume, and head-gesture Siri interactions. AirPods Max 2 also features a new high dynamic range amplifier, with wired connections supporting 24-bit, 48 kHz lossless audio; wireless audio latency has also been reduced. Notably, the first-generation AirPods Max was released on December 15, 2020. AirPods Max 2 comes in Midnight, Starlight, Orange, Purple, and Blue finishes. Pre-orders begin on March 25, with official availability in early next month, priced at RMB 3,999. Source

    NVIDIA introduces DLSS 5

    On March 16, NVIDIA announced DLSS 5, calling it the most significant breakthrough in computer graphics since the introduction of real-time ray tracing in 2018. DLSS 5 introduces a real-time neural rendering model that injects realistic lighting and materials into pixels, enabling developers to “deliver computer graphics previously only possible in Hollywood visual effects.” DLSS 5 uses an offline video AI model, taking each frame’s color and motion vectors as input, and integrates lighting and materials that are anchored to the source 3D content while remaining consistent across frames, enabling real-time gameplay at up to 4K resolution. Source

    Apple Watch in mainland China rolls out mobile atrial fibrillation (AFib) irregular rhythm notification feature

    On March 17, Apple Watch in mainland China officially rolled out the AFib History feature. Users diagnosed with atrial fibrillation can enable this feature to access key insights, such as the estimated frequency of AFib indications detected in their heart rhythm, helping them better understand their health. Users can also receive weekly notifications about their estimated AFib frequency and view detailed historical data in the Health app, including lifestyle factors that may impact AFib, such as sleep, alcohol consumption, and exercise. Source

    Zhipu releases GLM-5-Turbo

    On March 16, Zhipu released GLM-5-Turbo, which it claims is the first foundation model deeply optimized for OpenClaw scenarios, with targeted enhancements made during training to meet OpenClaw’s core requirements. The GLM-5-Turbo model strengthens key capabilities such as tool invocation, command tracking, persistent tasks, and long-chain execution, supporting a maximum output of 128K tokens and a context length of 200K. It also features thinking mode, streaming output, function calling, context caching, and MCP, ensuring greater stability and reliability for multi-step tasks and enabling OpenClaw to transition quickly from conversation to execution. In addition, the model is better suited to handle high data throughput and long logical-chain Lobster tasks, further improving execution efficiency and response stability, making it suitable for integration into real-world workflows. However, the model is currently still in an experimental stage and remains closed-source; related research outcomes and features will be integrated into Zhipu’s next open-source model. Source

    National Satellite Internet System and Service Standardization Technical Committee approved for establishment

    On March 16, the State Administration for Market Regulation officially announced the approval of the National Satellite Internet System and Service Standardization Technical Committee. In recent years, with the accelerated deployment of low-Earth orbit satellite constellations, satellite internet applications have continued to expand across areas such as direct-to-device connectivity, emergency communications, marine fisheries, transportation and logistics, and connectivity in remote regions. China’s satellite internet industry ecosystem has become increasingly active, with growing industrial capabilities. To fully leverage the guiding and foundational role of standardization in building new satellite internet infrastructure and to promote the healthy development of the intelligent economy, the newly established committee will address the rapid pace of technological advancement and the influx of diverse market participants by consolidating efforts across stakeholders, promoting resource sharing, accelerating the development of urgently needed standards such as terminology, in-orbit constellation operation evaluation, and key system products, continuously improving the satellite internet standards system, actively participating in international standard-setting, and driving the high-speed, standardized, and healthy development of the satellite internet industry. Source

    China Academy of Information and Communications Technology releases Blue Book on Next-Generation Smart Terminals (2025)

    On March 16, the China Academy of Information and Communications Technology (CAICT) officially released the Blue Book on Next-Generation Smart Terminals (2025) — From “AI + Terminals” to AI-Native Devices. The report systematically reviews the evolution of full-stack technology integration in smart terminals, analyzes four core characteristics of AI terminals—cognitive collaboration, scenario anticipation, intent-driven interaction, and service symbiosis—and points out that the intelligent interconnection of everything is driving the industry from standalone intelligence toward higher-level forms of collective intelligence and scenario-based service ecosystems. In light of current industry challenges such as ecosystem fragmentation and privacy concerns in device-cloud collaboration, the Blue Book recommends coordinated efforts across four dimensions: strengthening policy guidance, accelerating standard unification, advancing key technology breakthroughs, and building a full-stack secure and trustworthy ecosystem, in order to ensure the high-quality development of next-generation smart terminals and position them as a primary platform for implementing the “AI+” initiative. Source

    News Worth a Quick Look

    • According to Blue Whale Tech on March 16, the copyright dispute between Bambu Lab’s MakerWorld platform and POP MART IP has been resolved. Bambu Lab and POP MART reached an amicable settlement through negotiations, and all related content has been fully taken down. The MakerWorld model community, which previously contained infringing content, no longer hosts any Labubu-related model files. Source
    • Spotify began testing the “Music Taste Profile” feature on March 13. This feature serves as the core of Spotify’s recommendation algorithm system, influencing personalized content such as Discover Weekly, recommendations, and annual summaries. It will first roll out to Premium subscribers in New Zealand, allowing users to view listening data across music, podcasts, and audiobooks. Users can also directly edit their taste profiles and even fine-tune future recommendations using simple language commands—for example, reducing certain genres or minimizing specific moods in recommendations. Source
    • vivo and iQOO announced on March 16 adjustments to the suggested retail prices of some of their products. According to an official statement posted in vivo’s community, due to continued significant increases in global semiconductor and memory costs, and after careful evaluation, the company will adjust the suggested retail prices of certain products starting at 10:00 on March 18, 2026. Specific models and pricing will be subject to official product pages. Source
    • According to Z Finance, ByteDance has suspended its Doubao AI glasses project. Source
    • According to Reuters, Alibaba has established the Alibaba Token Hub business group, directly overseen by CEO Eddie Wu. The group includes Tongyi Lab, the MaaS business line, the Qwen business unit, the Wukong business unit, and the AI Innovation business unit. Source
    • On March 16, Apple acquired MotionVFX. Based in Poland and founded in 2009, MotionVFX is a well-known third-party plugin developer for Final Cut Pro, DaVinci Resolve, and Apple Motion. Source
    • According to Instagram’s Help Center, after May 8, 2026, Instagram will no longer support end-to-end encrypted messaging. Source
    • Sony has introduced an upgraded PSSR upscaling technology in the latest PS5 Pro update, improving image stability and enhancing detail clarity. Source
  • SSPAI Morning Brief: PlayStation Tests Personalized Game Pricing as China Approves Its First Brain–Computer Interface Medical Device

    SSPAI Morning Brief: PlayStation Tests Personalized Game Pricing as China Approves Its First Brain–Computer Interface Medical Device

    Morning Brief

    1. China’s market regulator cracks down on misleading ads with “big claims and tiny disclaimers”
    2. CCTV’s 3.15 Gala exposes “data poisoning” attacks on large AI models
    3. PlayStation Store tests dynamic pricing, with different prices for different users
    4. Adobe reaches settlement with the U.S. Department of Justice over deceptive subscription cancellation practices
    5. MacBook Neo rated the most repairable Apple laptop in 14 years
    6. China approves its first brain–computer interface medical device
    7. Other News Worth a Quick Look

    China’s market regulator cracks down on misleading ads with “big claims and tiny disclaimers”

    On March 13, China’s State Administration for Market Regulation (SAMR) issued a notice titled Notice on Strengthening the Regulation of Prompting Language in Advertisements, launching a six-month campaign to clean up related advertising irregularities.

    The notice requires:

    • Cracking down on misleading “big headline, tiny disclaimer” advertising. Ads that use hard-to-read small text to indicate restrictions or unfavorable information will be targeted; in serious cases they may be classified as false advertising;
    • Investigating cases where required prompting language is missing or intentionally weakened. Advertisements for pharmaceuticals, health supplements, and similar products must clearly display legally required warnings; AI-generated content must also be clearly labeled. Unauthorized use of another person’s likeness may be deemed false advertising;
    • Investigating advertisements that use absolute language. Terms such as “No.1” or “Best” will be strictly scrutinized. If unclear industry scope definitions lead to consumer misunderstanding, such ads may be classified as false advertising;
    • Strengthening oversight of cited claims and advertiser self-verification. Citations must indicate their sources; when advertisers use their own data as proof, timeframes, regions, and other limiting conditions must be clearly stated. Distorting data, presenting probabilistic conclusions as certainty, or omitting limiting conditions may all constitute false advertising;
    • Enhancing supervision of major advertising distribution channels. Internet platforms and outdoor advertising publishers must fulfill their responsibilities and ensure that disclaimers and prompt information are prominently displayed in advertising designs.

    CCTV’s 3.15 Gala exposes “data poisoning” attacks on large AI models

    On March 15, China Central Television’s annual “3.15 Consumer Rights Gala” exposed a black-market industry chain dedicated to “poisoning” AI large models and manipulating their outputs.

    The report explains that GEO originally refers to marketing optimization for generative AI search results, similar to SEO during the traditional search engine era. However, some illicit service providers are exploiting Generative Engine Optimization (GEO) techniques to flood the internet with fake promotional articles. These articles interfere with how AI models gather and cross-validate information, allowing companies to manipulate AI outputs so that advertisements or fabricated claims appear as the “standard answers” recommended by AI to consumers.

    During the program, industry insiders demonstrated the process. They randomly purchased a piece of software called the “Liqing GEO System” and fabricated a fictional smart wristband named Apollo9. The GEO software immediately generated promotional articles containing absurd claims such as “quantum entanglement sensors” and “black-hole-level battery life,” then automatically published them across multiple self-media accounts. Within two hours, major AI models had already scraped these posts. When users later asked for smart band recommendations, the AI repeated the fabricated promotional claims verbatim. After publishing additional fake product reviews generated by the GEO system, two AI services began recommending the wristband near the top of their responses within three days.

    The report revealed that the growing demand for GEO services has led to the emergence of numerous companies and platforms specializing in mass content publishing. These operations handle bulk posting requests, automatically releasing hundreds of machine-generated spam articles every day and earning tens of yuan per article.

    PlayStation Store tests dynamic pricing, with different prices for different users

    According to the PlayStation price tracking website PSprices, the PlayStation Store is conducting large-scale A/B testing of dynamic pricing. The testing has been quietly underway for more than four months and expanded in March to the United States—PlayStation’s largest market—covering over 70 regions worldwide across Europe, the Americas, and Asia, and involving more than 190 games. Japan is the only major market not included in the test.

    The testing was discovered through analysis of backend API data. Under the dynamic pricing system, players are randomly divided into groups. Some selected users see exclusive discounted prices lower than the standard retail price when browsing the same game, with discounts reaching up to 27.8%. Games included in the dynamic pricing test range from popular third-party titles by major publishers such as 2K and Rockstar to key first-party releases including Marvel’s Spider-Man 2, God of War: Ragnarök, and Helldivers 2.

    Data released so far suggests that Sony has created multiple test groups targeting different markets and user segments to evaluate various discount levels. This personalized pricing has also appeared in specific promotional events. For example, during a recent sale, ordinary players saw a 25% discount for Helldivers 2, while some users in experimental groups were offered discounts as high as 56%.

    Adobe reaches settlement with the U.S. Department of Justice over deceptive subscription cancellation practices

    On March 14, Adobe agreed to pay a $75 million fine and provide $75 million worth of free services to affected users to settle a lawsuit filed by the U.S. Department of Justice over allegations that it concealed subscription cancellation penalties.

    The lawsuit began in 2024, with the core dispute focusing on Adobe hiding expensive early termination fees within supplementary clauses or hyperlinks in user agreements. If users canceled an annual subscription plan midway, they could be charged 50% of the remaining subscription cost, often amounting to several hundred dollars. Adobe was also accused of creating complicated phone-based customer service processes that deliberately made cancellation more difficult. At the time, the Department of Justice argued that Adobe’s practices violated the Restore Online Shoppers’ Confidence Act, which is intended to prevent deceptive billing practices in online services.

    Under the settlement agreement, Adobe did not admit to any wrongdoing. The company has not yet clarified which users will qualify for the free service compensation or what form it will take, stating only that eligible users will receive notifications. Adobe also emphasized that it has improved its sales processes in recent years, making cancellation terms clearer at the time of purchase.

    The settlement is also seen as another instance of the Trump administration scaling back consumer protection lawsuits initiated during the Biden era. Previously, the Federal Trade Commission and the Department of Justice reached settlements in major cases involving Amazon Prime subscriptions and Ticketmaster’s ticketing services.

    MacBook Neo rated the most repairable Apple laptop in 14 years

    On March 13, iFixit released a teardown report for Apple’s budget laptop MacBook Neo, awarding it a repairability score of 6 out of 10 and calling it “the most repairable MacBook in 14 years.”

    The teardown revealed significant changes to the internal design. The most notable update is the complete abandonment of Apple’s long-used glued-in battery, replaced instead with a battery tray secured by 18 screws. This design greatly reduces the difficulty and risk of battery replacement. The move is believed to prepare Apple for upcoming European Union regulations set to take effect in 2027, which will require batteries in portable devices to be replaceable. Tests also showed that the device no longer imposes software restrictions on non-original parts. After replacing components such as the display, battery, or even the Touch ID fingerprint module, users only need to run Apple’s official calibration software for proper compatibility, without triggering warnings about non-genuine parts.

    Other internal components of the MacBook Neo also demonstrate a higher level of modular design. The USB-C ports and 3.5 mm headphone jack are separate modules, meaning they can be replaced individually without swapping out the entire motherboard. The disassembly process for the display antenna and hinges has also been significantly simplified. In the control area, the keyboard still requires removing 41 screws but has moved away from the previous riveted structure. Meanwhile, the trackpad has dropped its force-touch functionality, making it easier to repair and cheaper to replace.

    China approves its first brain–computer interface medical device

    According to Bloomberg, on March 13 China’s National Medical Products Administration approved the country’s first commercial invasive brain–computer interface (BCI) product. The device was developed by Neuracle and is designed to help adult patients with partial paralysis caused by spinal cord injuries improve hand grasping ability by controlling a robotic glove with their thoughts.

    The approved system integrates an implanted brain sensor, a robotic glove, signal decoding algorithms, and related medical software. It has previously been tested on 36 patients. Unlike devices such as Elon Musk’s Neuralink, which implant electrodes directly into the cerebral cortex, Neuracle’s sensor is placed outside the brain’s outermost membrane. This approach is relatively less invasive, but because it has fewer signal detection channels, the current system is limited to patients who still retain some upper-arm function. It cannot yet enable more advanced interactions, such as allowing fully paralyzed patients to browse the internet or play games.

    Brain–computer interfaces have been identified as one of six strategic emerging industries in China’s latest five-year plan. Authorities have pledged to streamline regulatory approval processes and are already working on guidelines for potential medical insurance reimbursement. Following news of the first commercial product approval, shares of related companies—including Sanbo Brain Science, Innovational Medical, and Nanjing Panda Electronics—have all surged by more than 10%.

    Investment and financing activity in the global brain–computer interface sector is also accelerating. Last week, another Chinese company competing in the Neuralink space, StairMed, completed a 500 million RMB funding round led by Alibaba and plans to conduct 40 additional human implant procedures this year. Chengdu-based Gestala, which is developing alternative BCI technologies, also announced a 150 million RMB seed round. In the United States, Science Corp recently raised $230 million to advance the commercialization of its implantable devices.

    Other News Worth a Quick Look

    • On March 14, 360 held a launch event in Beijing, unveiling the so-called “360 Secure Lobster” client and the “360 Secure Lobster Box” hardware terminal, along with a security product called “360 Lobster Guardian” designed specifically to address OpenClaw security issues. 360 claims this is the first OpenClaw agent product in China designed around a “security mode.” The company also promoted the solution as addressing what it describes as the four major obstacles hindering OpenClaw adoption—“difficult installation, difficult maintenance, instability, and security risks.” According to 360, the Secure Lobster product line provides a comprehensive “fully prepared out-of-the-box, all-around protection” solution aimed at allowing ordinary users to benefit from the technology and achieve what it calls “Lobster freedom.” Zhou Hongyi personally helped install and deploy “360 Secure Lobster” for users at the company’s headquarters campus.
    • According to The Information, ByteDance has paused the global release of its latest AI video generation model, Seedance 2.0, due to copyright disputes with major Hollywood studios. The model had originally been scheduled to open to global users in mid-March this year.
    • Mark Gurman reports that:
      • Apple is deeply adapting iOS 27 for a foldable iPhone, introducing split-screen multitasking to the iPhone for the first time. Because the foldable panel must be purchased from Samsung at a high price, combined with a complex hinge mechanism and premium materials, the device will be very expensive;
      • Apple’s first smart home device with a 7-inch display, originally scheduled to launch in March this year, has been delayed to September due to ongoing development delays in the new version of Siri and related AI features;
      • Liquid Glass will not undergo major changes in iOS 27. Although Alan Dye, the vice president who led the design, left with several key team members for Meta late last year, the interface has been in development for years and is deeply integrated into Apple’s ecosystem. His successor Steve Lemay will focus primarily on optimization;
      • Apple will not consider equipping the MacBook Neo with high-end components such as touch displays within the next three years.
  • SSPAI Morning Brief: Apple Cuts App Store Fees in China

    SSPAI Morning Brief: Apple Cuts App Store Fees in China

    Morning Brief

    1. Apple announces lower App Store commission rates in China
    2. Samsung adjusts low-level flashing mechanism for Galaxy devices
    3. PEGI to revise game age rating rules starting in June
    4. NVIDIA open-sources Nemotron 3 Super AI model weights
    5. Feishu launches official OpenClaw plugin
    6. Tencent’s WorkBuddy (Lobster edition) adds direct WeChat connectivity
    7. OpenAI CEO says AI faces multiple obstacles in the United States
    8. Microsoft to roll out Xbox FSE feature to all Windows 11 PCs
    9. Microsoft announces the return of the Xbox backward compatibility program
    10. Steam Machine Verified standards for gaming consoles published
    11. Firefox 115 ESR support extended to August 2026
    12. Other News Worth a Quick Look

    Apple announces lower App Store commission rates in China

    Apple announced on its developer website that, after discussions with Chinese regulators, the company has decided to adjust App Store distribution commission rates in mainland China. Specifically:

    The standard commission rate for in-app purchases and paid apps will be reduced from 30% to 25%;
    At the same time, for developers eligible under the App Store Small Business Program and the Mini Apps Partner Program, the commission rate for in-app purchases and the commission applied to auto-renewing subscriptions starting from the second year will be reduced from 15% to 12%.

    Apple stated that the new rates will take effect on March 15, 2026, and developers will not need to re-sign related agreements. The company also emphasized that it will continue to provide fair and transparent terms for developers distributing apps in China, and pledged that overall App Store commission levels in China will not exceed those in other markets. Source

    Samsung adjusts low-level flashing mechanism for Galaxy devices

    Tech outlet Android Authority reported on March 11 that Samsung has adjusted the low-level flashing mechanism for Galaxy devices in the latest One UI 8.5 firmware. Tools such as Odin—commonly used for flashing official firmware, installing custom ROMs, or recovering devices—and the associated Download Mode entry point were briefly believed to have been disabled. Affected models include the Galaxy Z Fold 7 and the newly released Galaxy S26 series. When users attempt to enter Download Mode, the traditional interface no longer appears; instead, a blank blue screen with an exit prompt is displayed.

    Further investigation shows that Download Mode and Odin still exist, but users must first enable the device’s Maintenance Mode before entering the relevant mode. Samsung has not yet released an official explanation for the change. Discussions generally suggest the adjustment may relate to strengthening system security, restricting firmware sideloading, and tightening ecosystem control. The change may also affect developers, repair services, and users who rely on Odin for device recovery or system downgrades. Source

    PEGI to revise game age rating rules starting in June

    The Pan European Game Information (PEGI) organization announced that it will update its game age rating standards starting in June. Under the new rules, games containing paid random items (such as card packs, gacha systems, or randomized reward keys) will receive a PEGI 16 rating, while social casino-style games will receive a PEGI 18 rating. The updated system will also consider factors such as in-game purchases, online communication features, and design elements that encourage frequent player return—such as daily missions or consecutive login rewards.

    For example, games featuring time-limited or quantity-limited purchase mechanics will typically receive a PEGI 12 rating; if spending features are disabled by default, the rating may be lowered to PEGI 7. If in-game purchases are tied to NFT assets that are required for gameplay and are tradable, the game will be rated PEGI 18.

    The new rules will apply only to games submitted for classification after June. PEGI said the changes reference recent reforms to the rating system of Germany’s Unterhaltungssoftware Selbstkontrolle (USK). The updated standards will place greater emphasis on online interaction risks and the influence of game design on consumer spending and gameplay behavior, rather than focusing solely on traditional factors such as violence or drug content.

    PEGI noted that after Germany updated its rating system, about 30% of games were affected by the new standards, with roughly one-third receiving higher age ratings. PEGI believes the updated rules will provide clearer information for parents to help them determine whether a game is appropriate for children, while also acknowledging that some game developers may question higher age classifications. Source

    In addition, Valve recently published a statement on the Steam support website responding to a lawsuit filed earlier by the New York State Attorney General regarding loot box systems in games such as Counter-Strike 2 and DOTA 2. Valve stated that paid loot box mechanics in games are similar to opening baseball card packs, blind boxes, or trading card packs in the real world, and therefore do not constitute gambling. Valve emphasized that players can fully enjoy these games without spending money, since items obtained from loot boxes are purely cosmetic and do not affect gameplay balance. The company also said its data shows that most players never open any loot boxes and simply play the games normally, meaning non-paying players are not at a disadvantage.

    Previously, the New York Attorney General filed the lawsuit in February, accusing Valve of “facilitating gambling behavior” through the Steam platform and popular online games, and seeking a permanent ban on promoting such mechanics along with financial penalties. In its statement, Valve said the company has long taken action against third-party gambling websites, banning more than one million Steam accounts involved in gambling, fraud, or theft, and introducing mechanisms such as trade reversal and trade cooldown systems to limit related activities. Source

    NVIDIA open-sources Nemotron 3 Super AI model weights

    On March 11, NVIDIA published a blog post announcing the release of Nemotron 3 Super, the company’s most powerful open-weight AI model to date.

    Nemotron 3 Super features 120 billion parameters and uses a Mixture-of-Experts (MoE) architecture, activating only about 12 billion parameters during inference. The model is designed for scenarios involving large-scale deployment of complex agentic AI systems. Combined with advanced reasoning capabilities, it helps autonomous agents complete tasks with higher accuracy. The model provides a context window of up to 1 million tokens, allowing agents to retain the full state of workflows in memory, avoid goal drift, and significantly reduce the cost of multi-step reasoning.

    NVIDIA stated that companies such as Perplexity, Palantir, and Siemens have already adopted the model in real-world applications, covering core workflows in areas including search, software development, telecommunications, and semiconductor design.

    NVIDIA has taken a highly open approach with Nemotron 3 Super. The company has released the model weights under a permissive license and published the full training methodology, including a training dataset exceeding 10 trillion tokens as well as evaluation recipes. Developers can currently access the model through Hugging Face and several cloud platforms, including Google Cloud and Oracle Cloud, with support for AWS and Azure coming soon. In addition, the model is available as an NVIDIA NIM microservice, allowing seamless deployment in on-premises data centers or cloud environments. Source

    Feishu launches official OpenClaw plugin

    Feishu announced the launch of an official OpenClaw plugin that allows OpenClaw to directly access collaboration tools within the Feishu ecosystem, including documents, group chats, calendars, tasks, and multi-dimensional tables. After granting plugin permissions, OpenClaw can read and analyze messages and document content within Feishu, while also performing actions such as writing documents, sending messages, creating schedules, and generating multi-dimensional tables. This reduces the need for users to repeatedly copy and paste content between AI tools and workplace software.

    The official plugin also supports message reading and replies, cloud document creation and updates, multi-dimensional table data management, calendar and participant management, and task management. These capabilities allow AI to better understand contextual information within collaborative workflows.

    Feishu stated that the goal of the plugin is to expand AI from a simple chat interface into an intelligent assistant capable of participating in real work tasks. However, the company also warned that because the plugin accesses work-related data such as messages, documents, and schedules, potential data security and privacy risks remain. In addition, since AI may produce misjudgments or “hallucinations,” users are advised to preview and confirm actions—such as sending messages or modifying content—before execution, rather than relying on full automation. Feishu also recommends testing the plugin within a personal account environment before gradually integrating it into real workplace settings. Source

    Tencent’s WorkBuddy (Lobster edition) adds direct WeChat connectivity

    Tencent’s WorkBuddy (Lobster edition) announced a new version early on March 12. The update introduces one-click WeChat connectivity, allowing users to remotely command the system through WeChat. The enterprise WeChat version also gains persistent connection capabilities along with automated task execution and monitoring features.

    Among the updates, the “one-click WeChat connection” feature stands out as particularly useful. To enable it, users open Claw Settings in the upper-right corner, select WeChat Customer Service Integration, and scan a QR code with their phone to complete the binding process. Once connected, a dedicated contact appears within WeChat. Tencent states that from that point forward, users can send simple text commands from anywhere, and WorkBuddy on their office computer will immediately execute tasks such as searching for information, conducting research, writing content, or processing files. All operations run locally on the computer. As long as the computer remains powered on, the system can operate continuously 24 hours a day, 7 days a week. Source

    OpenAI CEO says AI faces multiple obstacles in the United States

    At the BlackRock U.S. Infrastructure Summit held in Washington, Sam Altman said that as AI technology develops rapidly, negative public narratives have begun to emerge. For example, some critics argue that data center construction drives up electricity prices, while certain companies cite AI as a reason when announcing layoffs. The combination of these factors has gradually turned AI into a controversial topic in public discussions. In addition, debates between governments and technology companies over regulation and the boundaries of power remain unresolved, adding further uncertainty to the future of the AI industry.

    Multiple opinion polls indicate that the American public is becoming more cautious about AI. A survey by NBC News found that 57% of respondents believe the risks of AI outweigh its potential benefits. Research from the Pew Research Center shows that about half of U.S. adults say they feel “more concerned than excited” about the development of AI, a figure that has increased by 13 percentage points since 2021. Altman also said that the United States still leads in the global AI race, but if AI adoption in industry and the economy does not progress quickly enough, that advantage could weaken. He believes AI still holds strong potential to drive economic growth and may reshape how society operates in the future. Source

    Microsoft to roll out Xbox FSE feature to all Windows 11 PCs

    On March 12, Microsoft announced that the Xbox FSE full-screen experience (Xbox Full Screen Experience), originally introduced last June for handheld gaming devices, has been officially renamed “Xbox Mode” and will begin rolling out to all Windows 11 PCs starting next month.

    According to publicly available information, Microsoft’s FSE full-screen experience is similar to Steam’s Big Picture mode and aims to provide a more user-friendly interface for gaming PCs connected to large displays. The mode supports not only games from Microsoft’s own platform but also titles from Steam, Epic Games Store, GOG, Ubisoft, and Blizzard’s Battle.net.

    With this mode, players can switch between different gaming platforms using only a controller, eliminating the need to frequently navigate menus with a mouse or touchscreen. The interface has also been redesigned specifically for controller input. Meanwhile, the system reduces background processes and postpones non-essential tasks to improve overall gaming performance. Source

    Microsoft announces the return of the Xbox backward compatibility program

    On March 12, Microsoft announced at GDC 2026 that it will restart the Xbox Backwards Compatibility Program. The initiative aims to allow players to access more classic titles from the original Xbox and Xbox 360 on current and future Xbox platforms.

    Microsoft has not yet revealed the specific implementation details of the revived program and has only stated that it will return in some form later this year.

    Previous reports suggest that the new Xbox backward compatibility plan may support running classic games directly on devices. At the same time, Microsoft’s next-generation Xbox hardware is rumored to be a compact gaming PC running Windows 11 with an Xbox-style interface. Under this architecture, a plausible scenario is that Microsoft could release official emulators for original Xbox and Xbox 360 games on Windows 11, enabling broader backward compatibility support. Source

    Steam Machine Verified standards for gaming consoles published

    At GDC 2026, Valve unveiled the hardware verification requirements for the Verified label for the Steam Machine gaming console and the Steam Frame VR headset. The Verified label is a game compatibility certification system introduced by Valve for devices such as Steam Deck, Steam Machine, and Steam Frame, indicating that a game can run smoothly on the device without additional configuration, with optimized controls, clear display, and full controller support.

    For Steam Machine consoles, Valve has set a performance benchmark of maintaining 30 FPS at a native resolution of 1080p. Valve also confirmed that the certification standards for Steam Deck Verified and Steam Machine Verified are identical. All games that have already received the Steam Deck Verified label will automatically be compatible and will directly receive the Steam Machine Verified label as well. In addition, the certification includes several core hardware requirements, such as overall hardware performance reaching six times that of the Steam Deck and adopting the same controller input standards as the Steam Deck.

    For the Steam Frame VR headset, Valve stated that if a game runs through PC streaming, no additional verification is required. However, for standalone operation on the device itself, games must meet certain performance benchmarks. VR games must maintain a stable 90 FPS, while 2D games must reach 30 FPS at a resolution of 720p.

    Valve also recommends that developers optimize their games for SteamOS devices, such as enabling offline play after the first launch, improving performance on lower-end hardware, and enhancing support for controller input. Source

    Firefox 115 ESR support extended to August 2026

    On March 11, Mozilla announced that it will further extend the support cycle for Mozilla Firefox 115 ESR on older operating systems. The extension applies to systems such as Windows 7, Windows 8.1, and macOS 10.12–10.14, with the support period extended from the originally planned March 2026 to August 2026.

    Mozilla also emphasized on its support page that the team plans to conduct another evaluation in July 2026 and will announce the final end-of-support schedule afterward. Therefore, August 2026 may not necessarily be the final deadline, as the timeline will continue to be adjusted based on user data and usage patterns. Source

    Other News Worth a Quick Look

    • Mark Gurman reports that Apple’s upcoming foldable iPhone may adopt an interface layout similar to the iPad, allowing users to view two apps simultaneously to enhance multitasking. The outer display is expected to be close in size to a smaller iPhone, while the inner display may use a wider aspect ratio similar to Google’s first-generation Pixel Fold rather than the proportions used by some existing foldable flagships. The device is not expected to run current iPad apps directly. However, Apple plans to introduce a sidebar design for its “core” apps and will provide developers with related tools so that iPhone apps can deliver an experience closer to iPad apps when displayed on a larger screen. The report also notes that because the device’s front panel will be extremely thin, Apple may not use Face ID. Instead, it may integrate Touch ID into the side power button and adopt a hole-punch front camera design, while still supporting the Dynamic Island feature. The foldable iPhone is expected to feature two rear cameras and may be priced at around $2,000. Source
    • DJI has recently begun teasing its first 8K flagship panoramic drone, the DJI Avata 360, and confirmed that the new product will launch on March 26. Official teaser posters indicate that the drone continues the Avata series’ integrated propeller guard design and familiar body structure, while potentially offering improved control and flight experience along with panoramic aerial shooting capabilities. Previous leaks suggest that the DJI Avata 360 uses a gray ducted-frame structure with fully enclosed spherical propeller guards. The gimbal reportedly supports a rotating storage design that protects the camera from scratches caused by gravel during takeoff and landing. Rumors also indicate the drone will feature a 1-inch sensor, support panoramic imaging, record up to 6K 60FPS HDR video, and include omnidirectional obstacle avoidance. Source
  • SSPAI Morning Brief: Google Launches Gemini Embedding 2 as ComfyUI Introduces App-Style Workflows

    SSPAI Morning Brief: Google Launches Gemini Embedding 2 as ComfyUI Introduces App-Style Workflows

    Morning Brief

    1. Google releases Gemini Embedding 2 model
    2. MediaTek unveils multiple IoT SoC platforms
    3. ComfyUI adds support for app-style workflows
    4. RØDE launches Video Core
    5. iReader releases iReader Neo 3 Ultra

    Google releases Gemini Embedding 2 model

    On March 11, Google announced the new Gemini Embedding 2 model. It is Google’s first native multimodal embedding model, capable of mapping text, images, videos, and documents into the same embedding space.

    Gemini Embedding 2 supports text, images, video, audio, and documents, and can recognize semantic intent across 100 languages. The new model simplifies complex data processing pipelines while enhancing multimodal application capabilities. Use cases include retrieval-augmented generation (RAG), semantic search, sentiment analysis, and data clustering. Gemini Embedding 2 can also accept multiple input types—such as “image plus text”—within a single request, enabling analysis of relationships across different media formats.

    The model is currently available in public preview through the Gemini API and Vertex AI. Source

    MediaTek unveils multiple IoT SoC platforms

    On March 9, MediaTek unveiled a series of IoT SoC platforms at Embedded World 2026 in Nuremberg, Germany, including the high-end Genio Pro solution as well as the lower-positioned Genio 420 and Genio 360.

    Genio Pro is built on TSMC’s advanced 3nm process and features CPU and GPU specifications similar to the mobile Dimensity 9400. It is designed to operate under industrial wide-temperature environments. The platform supports up to 16 cameras and as many as three 4K displays, and is compatible with a range of Linux distributions as well as the open-source ROS robotics operating system.

    Genio 420 uses a 6nm process and provides 7.2 TOPS of system AI computing power while integrating 16GB LPDDR5X memory. It remains pin-compatible with the existing Genio 720 and Genio 520, allowing users to migrate between chip models more easily.

    The lower-tier Genio 360 and Genio 360P deliver system AI performance of 6 TOPS and 8.5 TOPS respectively. They support up to 8GB LPDDR4X-3733 memory and can meet the requirements for running 2B parameter models on-device. Source

    ComfyUI adds support for app-style workflows

    On March 10, ComfyUI introduced three new features: App Mode, App Builder, and ComfyHub. The goal is to transform the platform’s node-based workflows—originally designed for advanced users—into shareable, runnable application interfaces, allowing people unfamiliar with node graphs to use ComfyUI workflows.

    App Mode enables any ComfyUI workflow to switch into a simplified interface with one click. The underlying node graph remains hidden, leaving only the inputs and outputs that end users need to operate. The workflow itself does not change and continues running on the same ComfyUI instance, backend, and task queue. As a result, existing nodes, model updates, and custom extensions can be inherited directly in App Mode.

    The newly introduced App Builder allows creators to configure which node parameters are exposed to users. Developers can decide which inputs and outputs appear in the interface, while more complex parameters remain inside the node graph. ComfyUI explains that a typical text-to-image workflow may include numerous parameters such as samplers, steps, CFG, schedulers, LoRA weights, seeds, and resolution, but only a small subset may need to be exposed in practical applications.

    ComfyUI also launched a preview version of ComfyHub, a public distribution hub for workflows and applications where users can browse and run community creations directly. These features are currently available on Comfy Cloud and Comfy Local. Source

    RØDE Launches Video Core

    On March 10, RØDE introduced the Caster Video Core, an integrated audio and video production solution designed for livestreaming and video capture scenarios.

    The device features an octa-core CPU and can switch between up to four video sources and five custom scenes. It includes three 1080p HDMI inputs, supports automatic frame rate switching, and provides two USB-C ports for connecting video or audio devices. It also comes with two Neutrik combo inputs, allowing users to connect XLR microphones, musical instruments, or other audio equipment.

    The product supports audio adjustments such as EQ, compression, filtering, and de-essing. It can also connect to mixing consoles such as the Caster Pro II or Duo, enabling additional audio and video input and output options. The USB-C ports can connect two computers simultaneously, or a portable storage drive to save recorded footage directly, and support dual-system livestreaming. Source

    Product appearance image, image from the news source

    iReader Launches iReader Neo 3 Ultra

    On March 10, iReader introduced its new flagship e-reader, the iReader Neo 3 Ultra. The device is powered by an octa-core processor, delivering a 200% overall performance improvement compared with previous quad-core chips.

    In terms of storage, the device comes with 4GB of RAM and 64GB of internal storage. The iReader Neo 3 Ultra features a new “Ultra Refresh” technology that increases frame rate while maintaining similar power consumption, enabling smoother page transitions and stable video playback. Combined with iClean 2.0 technology, the device uses AI-based image recognition algorithms to introduce an improved partial refresh mechanism, effectively reducing ghosting and flickering while enhancing reading comfort.

    For the display, the Neo 3 Ultra is equipped with a 300 PPI recessed screen made with Carta 1300 material, offering sharper text edges and clearer image details. The device measures just 7.5mm thick and weighs about 150g. It also includes a front light with 30 levels of brightness and color temperature adjustment. With ComfortLight Pro+ technology, the lighting remains soft and evenly distributed, allowing fine tuning based on ambient light conditions and personal preference.

    The device includes a 2000mAh battery, rated for up to 53 hours of reading or 18 days of standby. The product is now available on e-commerce platforms with a price of 1,499 yuan. Source

    Product appearance image, image from iReader
  • SSPAI Morning Brief: Gemini Comes to Google Workspace, ChatGPT Gains Shazam Music Recognition

    SSPAI Morning Brief: Gemini Comes to Google Workspace, ChatGPT Gains Shazam Music Recognition

    Morning Brief

    1. Honor Magic V6 unveiled
    2. OPPO announces price increases for selected models
    3. Sonos releases portable speaker Play and Era 100 SL
    4. Intel introduces Core (2nd Gen) performance-core Bartlett Lake 12P processor
    5. Meta acquires Moltbook
    6. Google Workspace adds new Gemini AI features
    7. ChatGPT introduces Shazam-style music recognition
    8. Adobe launches Photoshop AI assistant
    9. Short News Worth a Look

    Honor Magic V6 unveiled

    On March 10, Honor officially launched its foldable flagship smartphone Magic V6 in China, available in four color options: Chitu Red, Sunrise Gold, Snowfield White, and Velvet Black. The device measures 8.75 mm thick when folded and 4.0 mm when unfolded, with a total weight of 219 g. The display features Honor’s self-developed “Ultimate Black Diamond” screen technology, with the outer display using a 5600-layer silicon nitride stacking structure, while the inner display incorporates ultra-tough UTG glass that improves crease resistance by 44%. Both displays have received TÜV Rheinland global eye-care certification. At its core, the device is powered by the full-performance eight-core Snapdragon 8 Gen 5 Elite Edition processor and comes standard with 120 W GaN fast charging. It also debuts the new-generation Qinghai Lake Blade Battery with a capacity of 7,150 mAh, featuring 32% silicon content and an energy density of 985 Wh/L. The imaging system includes a 64 MP periscope telephoto camera with a 1/2-inch sensor and CIPA 6.5-level optical stabilization. The device uses a 2800 MPa “Shield Tunnel Steel” hinge structure and supports IP68 and IP69 dust and water resistance, as well as BeiDou satellite messaging. On the software side, it runs MagicOS 10 and supports Mac Tap-to-Transfer as well as compatibility with Apple’s iWork suite.

    Honor Magic V6 will be available in four configurations: 12 GB + 256 GB, 12 GB + 512 GB, 16 GB + 512 GB, and 16 GB + 1 TB. Prices start at 8,999 yuan, with official sales beginning at 10:08 on March 13. Source

    OPPO announces price increases for selected models

    On March 10, OPPO announced that due to rising costs and supply shortages of key storage components such as DRAM and NAND flash, it will adjust the official retail prices of some existing products starting at 00:00 on March 16. The price adjustments will affect devices in the OPPO A series, K series, and certain OnePlus models. However, the OPPO Find series, Reno series, and OPPO Pad tablets will not be included in this round of price changes. Source

    Sonos releases portable speaker Play and Era 100 SL

    On March 10, Sonos announced the release of the Sonos Play, a mid-sized portable speaker positioned between the Roam 2 and Move 2, as well as the Era 100 SL, a microphone-free version of the Era 100 speaker. The Sonos Play weighs approximately 3 pounds and measures 7.6 × 4.4 × 3 inches. Its acoustic structure includes two tweeters positioned at 90-degree angles, a mid-woofer, and two newly added passive radiators. It supports Wi-Fi, Bluetooth, AirPlay 2, and Spotify Connect connectivity, and carries an IP67 rating for water and dust resistance. The speaker also introduces Bluetooth multi-speaker networking (compatible with Move 2), a removable hanging loop design, and a built-in USB-C port that supports line-in, Ethernet expansion, and reverse charging. It comes with a wireless charging base, though a power adapter is not included. The Era 100 SL retains the full acoustic architecture of the Era 100 but removes the microphone component, targeting the entry-level market.

    The Sonos Play is priced at $299, while the Era 100 SL is priced at $189. Both products are available for pre-order starting today and are scheduled to ship on March 31. Source

    Intel introduces Core (2nd Gen) performance-core Bartlett Lake 12P processor

    On March 9, Intel officially introduced the Core (2nd Gen) performance-core processor codenamed Bartlett Lake 12P, designed for embedded and edge workloads. The processor lineup offers up to 12 P-cores with no E-cores. The series is built on Intel 7 process technology and includes three TDP configurations of 125W, 65W, and 45W. Maximum single-core turbo frequency reaches up to 5.9 GHz, while the all-core turbo frequency can reach up to 5.3 GHz. Intel claims the processor outperforms AMD’s Ryzen 7 9700X in metrics such as maximum PCIe latency, deterministic response time, and deterministic performance. It supports H610 (E), Q670 (E), R680E, and W680 motherboards, with memory overclocking available on the W680 platform. Source

    Meta acquires Moltbook

    On March 10, Meta announced the acquisition of Moltbook, a Reddit-like social network founded by Matt Schlicht and Ben Parr, and will integrate it into Meta’s Superintelligence Lab. Moltbook is a social platform built on OpenClaw. After OpenClaw gained popularity, the platform drew attention for branding itself as “Reddit for agents.” It also sparked public discussion with topics such as agent-encrypted language and agent religions. However, security research institutions later confirmed that the platform had significant security vulnerabilities, including long-exposed unencrypted database credentials and the ability for human users to impersonate AI agents to post content.

    Meta CTO Andrew Bosworth previously stated that compared with AI agents simulating human conversation, Meta is more interested in understanding how humans intervene in such networks through system vulnerabilities. The financial details and terms of the acquisition have not yet been disclosed. Source

    Google Workspace adds new Gemini AI features

    On March 10, Google announced a range of Gemini-powered AI features for Docs, Sheets, Slides, and Drive. In Docs, a new “Help me create” tool allows users to pull information from Gmail, Chat, and Drive to generate document drafts. It also introduces “match writing style” and “match format” options to standardize tone or fill in details such as airline and hotel information based on templates. Sheets can now extract data across platforms from a single instruction—such as organizing a moving checklist—and generate formatted tables. It also introduces a “Fill with Gemini” feature that can automatically categorize data or pull real-time information from Google Search to complete spreadsheets. Slides can generate a single slide aligned with the presentation theme using context from files, emails, and web pages, with plans to support generating full presentations from a single prompt in the future. Drive adds “AI Overview” and “Chat with Gemini,” allowing users to search files using natural language, receive summaries, or obtain detailed answers to complex questions—such as tax advice—based on private data across documents, emails, and calendars.

    These features are now available in English beta for global Google AI Ultra and Pro subscribers (Drive functionality is currently limited to users in the United States). Source

    ChatGPT introduces Shazam-style music recognition

    On March 10, OpenAI announced a partnership with Shazam to add music recognition functionality to ChatGPT. After manually searching for and adding Shazam in the Apps section within the ChatGPT client settings, users can trigger the feature with prompts such as “Shazam, what song is playing right now?” This will open the web version of Shazam inside ChatGPT. Once the music is identified, Shazam provides a music card in the reply box, allowing users to recognize songs and preview audio clips without leaving the ChatGPT app. Source

    Adobe launches Photoshop AI assistant

    On March 10, Adobe announced that the Photoshop AI Assistant has entered beta testing on both web and mobile platforms, alongside updates to its Firefly media creation tools. The Photoshop AI Assistant allows users to execute Photoshop editing commands through natural language instructions. Paid users can enjoy unlimited generation until April 9, while free users receive an initial quota of 20 generations. Meanwhile, Firefly introduces an experimental “AI Markup” feature, enabling users to trigger AI-driven object transformations or background modifications through simple hand-drawn annotations. Source

    Short News Worth a Look

    • On March 10, Zhipu announced the launch of AutoClaw, which comes with more than 50 popular prebuilt Skills and supports one-click integration with instant messaging tools such as Feishu. The tool supports both macOS and Windows platforms and offers a certain amount of free Zhipu usage quota along with third-party Coding Plan and API support. Source
    • According to Bloomberg, citing Mark Gurman, the long-rumored Apple smart home display has been delayed until September this year. Apple is reportedly still waiting for the final delivery of the new version of Siri in order to provide a complete smart ecosystem experience for the upcoming product. Source
    • On March 10, a Capcom director announced via a video message that story DLC for the upcoming title Resident Evil 9: Requiem is currently in development. In addition to continued performance bug fixes, future updates will also introduce a Photo Mode and mini-games as part of additional content updates. Source