SSPAI

Tag: Cloudflare

  • SSPAI Morning Brief: Cloudflare Acquires Vite Creator Evan You’s VoidZero to Accelerate AI-Powered Web Development

    SSPAI Morning Brief: Cloudflare Acquires Vite Creator Evan You’s VoidZero to Accelerate AI-Powered Web Development

    Morning Brief

    1. Passenger Cars Keep Getting Heavier, Gaining Nearly 400 kg Over 12 Years
    2. iFixit Tears Down Counterfeit Apple Products from Huaqiangbei
    3. Cloudflare Acquires the Company Behind Vite
    4. rsync Sparks Controversy After an AI-Assisted Maintenance Bug
    5. OpenAI Discusses Equity Donation With the U.S. Government
    6. Average Vehicle Age on U.S. Roads Reaches 13 Years
    7. News Worth a Quick Look

    Passenger Cars Keep Getting Heavier, Gaining Nearly 400 kg Over 12 Years

    According to a report by CCTV, passenger cars on the market have been collectively “bulking up” in recent years. Data shows that the average curb weight of new passenger vehicles in China reached 1,704 kg in 2024, nearly 400 kg heavier than the 1,312 kg average recorded in 2012. Moreover, the pace of weight gain continues to accelerate. Between 2020 and 2024 alone, the increase in vehicle weight exceeded the total gain accumulated during the eight-year period from 2012 to 2020. New energy vehicles have become the primary contributors to this trend, with many mainstream models weighing more than two metric tons and some reaching as much as 3.8 tons—heavier than a light-duty truck.

    Experts attribute the increase largely to the automotive industry’s transition toward electrification and consumers’ growing demand for more features. As the industry remains in a rapid phase of EV adoption, components such as traction batteries add significant weight. In today’s mainstream household EVs, battery packs typically weigh between 500 and 650 kg, while long-range models equipped with larger batteries can reach 700 to 800 kg. At the same time, electric motors, power control systems, and battery protection structures integrated into the chassis further increase overall vehicle mass.

    The report notes that excessive vehicle weight reduces efficiency, accelerates tire and brake wear, and can lead to handling issues such as understeer. Energy consumption also rises significantly. For example, a three-ton battery-electric SUV typically consumes more than 20 kWh per 100 kilometers. Industry estimates suggest that reducing vehicle weight by 100 kg can lower energy consumption by approximately 7.5% per 100 kilometers. Additionally, every 20% increase in vehicle weight raises road surface damage rates to roughly 2.07 times the original level, increasing urban road maintenance costs.

    Policymakers have already begun addressing the issue. China’s national standard Limits of Energy Consumption for Electric Vehicles, Part 1: Passenger Cars came into effect on January 1 this year. Through stricter and more systematic performance requirements, the regulation pushes automakers toward meaningful technological upgrades and lower energy consumption. New purchase tax incentive standards set to take effect in 2026 stipulate that battery-electric vehicles with a curb weight exceeding 2,710 kg must achieve CLTC energy consumption of no more than 19.1 kWh per 100 kilometers to qualify for the 50% purchase tax reduction. This measure is expected to curb unrestricted vehicle weight growth to some extent.

    iFixit Tears Down Counterfeit Apple Products from Huaqiangbei

    Recently, iFixit released a video showcasing the teardown of counterfeit Apple products purchased from Shenzhen’s Huaqiangbei electronics market, including fake versions of the Apple Watch Ultra 3, AirPods Max 2, and AirPods Pro 3. All three products together cost less than $100.

    The counterfeit Apple Watch Ultra 3 came preloaded with apps such as Alipay, but many built-in icons—including Compass and Weather—had no actual functionality, merely toggling Night Mode when tapped. On the rear casing, a plastic shell decorated with painted fake sensors replaced the genuine ceramic and sapphire sensor assembly found on the authentic product. The display assembly method also differed significantly from Apple’s design. Inside, the fake watch used a 260 mAh lithium-polymer pouch battery, compared with the genuine model’s 542 mAh battery, and it was directly soldered to the motherboard.

    The counterfeit AirPods Max 2 replaced aluminum components with low-quality plastic. The headband exhibited unusual bending, while the ear cushions appeared prone to rapid wear. Although the device could reproduce a convincing Apple-style pairing animation, it lacked active noise cancellation entirely. Internally, the wiring featured crude soldering work, and steel weights fixed with resin had been added inside the ear cups to imitate the heft of the genuine headphones.

    The counterfeit AirPods Pro 3 used silicone ear tips glued directly onto the earbuds. Inside were pouch batteries and inexpensive soldered copper wires instead of proper flex cables. Apple’s force-sensor controls were replaced with physical buttons, and the counterfeit version lacked both heart-rate sensors and noise-canceling microphones.

    According to iFixit, these counterfeit products not only fail to deliver any of the advanced features they attempt to imitate, but also pose potential fire hazards due to their use of unbranded and uncertified batteries. Built with low-quality components and offering virtually no repairability, they are likely to become electronic waste in a very short time.

    Cloudflare Acquires the Company Behind Vite

    On June 4, Cloudflare announced its acquisition of open-source JavaScript tooling company VoidZero. Core products under the VoidZero umbrella—including the Vite build tool, Vitest testing framework, Rolldown, and Oxc—will be integrated into the Cloudflare Workers platform.

    VoidZero was founded in 2024 by Chinese developer Evan You, creator of the Vue.js framework and Vite. Following the acquisition, You and his team will join Cloudflare’s Emerging Technologies and Incubation (ETI) division. Cloudflare CEO Matthew Prince stated that as AI-assisted code generation continues to grow rapidly, development tools must become lighter and more efficient. The integration aims to provide developers and AI agents with the fastest path from local code creation to deployment across Cloudflare’s global network.

    Addressing concerns from developers about the future of the ecosystem, Cloudflare pledged that Vite and its related tooling would remain vendor-neutral, continue to be released under the MIT License, and stay community-driven. The company also announced a $1 million independent Vite Ecosystem Fund dedicated to supporting open-source contributors and maintainers outside both the Cloudflare and VoidZero organizations.

    According to official figures, Vite currently surpasses 130 million weekly downloads worldwide. Cloudflare’s Vite plugin alone accounts for approximately 13.9 million weekly downloads, representing more than 10% of the total. Financial details of the acquisition have not been disclosed.

    rsync Sparks Controversy After an AI-Assisted Maintenance Bug

    According to The Register, a recent backup failure in a new version of the veteran open-source file synchronization tool rsync has triggered a broader debate about the use of AI in maintaining critical open-source software. Originally created in the 1990s, rsync has become a widely used backup and synchronization component throughout the Unix and Linux ecosystem, with countless enterprise IT departments, NAS devices, and backend scripts relying heavily on its stability.

    After the release of rsync 3.4.3, some users reported that incremental backups were no longer functioning correctly. While investigating the bug, users discovered references to Claude in the project’s commit history, prompting widespread criticism. One developer angrily posted on GitHub, telling the maintainers to “do not vibe fuck up this software,” a comment that quickly received thousands of reactions in agreement.

    In response, rsync creator Andrew Tridgell, a veteran software engineer with 40 years of experience, published an article defending his approach. He acknowledged that the new version did contain a bug affecting certain backup workflows and apologized for the issue, but firmly rejected accusations that he had blindly accepted AI-generated code. Tridgell explained that he primarily uses tools such as Claude for what he described as “grunt work,” including rewriting legacy test suites in Python to strengthen security testing, and emphasized that all generated code undergoes rigorous manual review by himself.

    Tridgell also pointed out that the software maintenance landscape has changed dramatically. Open-source maintainers are now facing a flood of security vulnerability reports, many of which are themselves generated by AI. To cope with the resulting increase in workload, he stated that he has no intention of abandoning AI-assisted development and plans to continue using such tools in the upcoming rsync 3.5 release, which will focus heavily on security improvements.

    OpenAI Discusses Equity Donation With the U.S. Government

    According to CNBC, U.S. President Donald Trump and OpenAI CEO Sam Altman are discussing a proposal that would allow the U.S. government to obtain an equity stake in the company. Negotiations have reportedly been ongoing for more than a year, with OpenAI potentially donating a portion of its shares to establish an investment vehicle similar to a “public wealth fund.”

    The idea was first proposed by Altman to the Trump administration in 2025, with the goal of allowing the American public to directly participate in and benefit from the economic gains generated by advances in artificial intelligence. Trump confirmed the concept to reporters aboard Air Force One on June 5, saying that he is considering arrangements under which certain AI companies would contribute equity so that “the American people become, in effect, partners.” He also revealed plans to meet with executives from several AI companies in the near future. Altman reportedly held extensive discussions with lawmakers and government officials in Washington last week.

    The negotiations come amid growing political interest in how AI-generated wealth should be distributed. Senator Bernie Sanders, for example, previously proposed a plan to impose a one-time 50% equity levy on major AI companies to create a sovereign wealth fund, and recently discussed the idea directly with Altman. During Trump’s second term, the administration has pursued a variety of technology and industrial policies and has already acquired equity stakes in companies including Intel, IBM, as well as several quantum computing and critical minerals firms.

    At the same time, the White House is increasing regulatory oversight of the AI industry. Trump recently signed two executive orders: one directing national security agencies to accelerate the adoption of advanced AI models, and another requiring AI companies to provide government access to major models 30 days before their public release.

    Average Vehicle Age on U.S. Roads Reaches 13 Years

    According to The Wall Street Journal, the average age of vehicles currently on U.S. roads has reached a record 13 years, up 10% from a decade ago. Driven by high new-car prices and economic uncertainty, this trend is reshaping the American automotive industry, pushing automakers and dealerships to focus more on repairs and used-car sales rather than new vehicle purchases.

    The average price of a new vehicle in the United States is now approaching $50,000 (approximately RMB 338,000). Combined with elevated auto loan interest rates, the cost of ownership has become increasingly difficult for consumers who depend on cars for daily transportation. Supply-chain disruptions since the pandemic, along with the recent cooling of the U.S. electric vehicle market, have further encouraged consumers to delay purchases. At the same time, advances in vehicle materials and safety technologies have made modern cars capable of remaining on the road much longer than before.

    As weak new-car sales become the norm, automakers and dealers are increasingly turning their attention to the existing vehicle market. Maintenance and repair operations now account for roughly half of the gross profit generated by U.S. dealerships. Traditional manufacturers such as Ford and General Motors are expanding their presence in the after-sales market. Ford, for example, recently introduced mobile repair services that send technicians directly to customers and uses connected-vehicle data to pre-position replacement parts before a breakdown occurs. Major automakers are also steadily expanding their certified pre-owned vehicle programs.

    Meanwhile, competition in the automotive aftermarket is becoming increasingly intense. Long wait times and a lack of pricing transparency at authorized dealerships have driven more American car owners—including those whose vehicles remain under warranty—to independent repair shops and nationwide quick-service chains. In response, large dealership groups such as Penske are reducing showroom space to create additional service bays and introducing customer perks such as complimentary cappuccinos, Wi-Fi-equipped workspaces, and one-hour express maintenance services in an effort to win customers back.

    News Worth a Quick Look

    • On June 7, leaker Sonny Dickson shared new photos of a foldable iPhone dummy unit, claiming it represents the finalized exterior design. He also suggested that the device may be available in only a single color option: white. Based on the model, the outer display adopts an edge-curved full-screen design. The rear microphone has been upgraded to a five-hole configuration, with the flash positioned beneath it. Meanwhile, the front-facing camera on the inner display has been moved to the upper-left corner.
    • Mark Gurman reports that Apple held a highly confidential internal meeting in early 2025 to address the underwhelming reception of Apple Intelligence and delays affecting the next-generation Siri. Senior executives reportedly realized that the company had fallen significantly behind in the AI race, prompting a major strategic shift. Following the reorganization, Mike Rockwell, who previously led the development of Vision Pro, took over responsibility for Siri, while former AI chief John Giannandrea departed the company earlier this year. After assuming the role, Rockwell quickly restructured the team and, together with software engineering chief Craig Federighi and others, helped facilitate deeper cooperation between Apple and Google. The situation also changed Tim Cook’s management style, with the CEO becoming unusually hands-on in shaping Apple’s AI roadmap.
    • Valve has recently rolled out a major update to the Steam Store homepage. The redesigned storefront features larger and more prominent images, along with more detailed descriptions. A new “Quick Browse” feature allows users to preview recommended games without opening individual store pages. Wishlist and DLC sections, which previously appeared only during sales events and promotions, are now permanently available on the homepage. The update also introduces a “Personalized Calendar” that recommends upcoming releases based on each user’s interests.
  • OpenClaw: After Two Weeks of Heavy Use, This AI Tool Has Completely Reshaped My Workflow

    OpenClaw: After Two Weeks of Heavy Use, This AI Tool Has Completely Reshaped My Workflow

    I believe many people have been repeatedly exposed over the past month to an AI tool called OpenClaw. According to social media posts and early community discussions, the tool went through several name changes before finally settling on OpenClaw. During the renaming period, there was even an incident where its social media accounts were maliciously squatted and used to launch a meme coin called $CLAWD to “harvest” users—adding a dramatic twist to the birth of OpenClaw.

    For a typical mobile-internet-era user like me, my instinctive reaction when encountering a problem has always been to look for yet another app to solve it. After more than two weeks of using OpenClaw, however, I noticed that my habits had quietly changed. Step by step, as I rebuilt my workflow around OpenClaw, it gradually became my first choice for handling digital-life needs.

    In this article, I’ll start by explaining what OpenClaw is and how to install it, then share some of my own usage scenarios and tips, hoping to offer some practical help to anyone interested in trying it out.

    What Is OpenClaw?

    OpenClaw is neither a simple large language model nor just a coding CLI. Instead, it is a locally running “digital life hub.” It can act as a personal AI assistant, and it can also take on parts of certain roles within a company. Its capabilities can be extended through Skills and tool integrations (including MCP scenarios), and by running a Gateway on a Mac mini or server, it enables cross-platform access and asynchronous scheduling. No matter where you are, you can interact with it in real time via various instant messaging tools.

    At the same time, all memories, file indexes, and personal habits generated within OpenClaw are stored in your own local workspace (such as Memory.md files, index databases, and Skills scripts). In a local deployment scenario, this gives you much greater control over your data and makes permission management and backups easier.

    How Do You Install OpenClaw?

    There are many options for hosting OpenClaw. At the moment, the most popular choice is the Mac mini, mainly for three reasons. First, it supports 24/7 continuous operation with low power consumption. Second, within the Apple ecosystem it can easily integrate Skills tied to Apple Reminders, Apple Notes, Apple Calendar, and more. Third, the hardware itself offers strong value for money. The 16GB + 256GB configuration often fluctuates in price on e-commerce platforms, but at the time of writing, many channels are approaching the 3,000 RMB range. Combined with the M4 chip and a minimum of 16GB unified memory, performance is more than sufficient.

    Beyond the Mac mini, OpenClaw can also be installed on devices running Windows, Linux, and other operating systems. If you don’t want to deploy OpenClaw locally, major cloud providers—such as Tencent Cloud, Alibaba Cloud, Cloudflare, and DigitalOcean—offer dedicated server images that support quick OpenClaw deployment as well.

    Overall, OpenClaw provides four installation methods: One-liner, npm, Hackable, and a macOS client. The One-liner supports macOS, Linux, and Windows; npm can be used via either npm or pnpm; Hackable comes in installer and pnpm variants. Each distribution method corresponds to different terminal commands, and you can choose whichever best suits your needs.

    After installation, OpenClaw automatically runs the openclaw onboard --install-daemon command for initial setup. In most cases, the default options are sufficient. The main configuration steps you’ll need to pay attention to are four key areas:

    The first is selecting an AI service provider and the specific model you want to use. Most providers offer both OAuth authorization and direct API key input, corresponding to different billing methods.

    Taking Google Gemini as an example: if you choose OAuth authentication, you can use services like Google Gemini 3 Pro, Google Gemini 3 Flash, Claude Opus 4.5, and Claude Sonnet 4.5 via Antigravity, with quotas refreshed every five hours. If you choose the Google API instead, you’ll need to create an API key in Google AI Studio or Vertex, and billing will be based on token and prompt usage. Once provider verification is complete, you can select your preferred model from the available list.

    The second step is configuring the communication Channels used to interact with OpenClaw. If you choose certain overseas messaging tools, issues may arise during setup, potentially causing the Gateway to fail to start. In that case, you have three options: let it fail and complete the setup first, then ask the AI to help fix it later; skip Channel configuration and move on; or choose a domestic messaging tool such as Feishu.

    The third step is selecting and installing Skills. Use the arrow keys to navigate, the spacebar to select or deselect, and press Enter to install. If you don’t want to spend too much time here, you can skip this step and install Skills later.

    The final step is launching the Gateway and choosing a control interface, such as TUI or Web UI. At this point, the initial OpenClaw configuration is complete. If you’re a macOS user, you can also install the official Companion App. This menu-bar utility makes it easy to adjust OpenClaw settings, monitor Gateway status, and even chat directly with OpenClaw.

    How Should You Choose an OpenClaw Model?

    As of the time of writing, OpenAI has just released ChatGPT 5.3-Codex, while Anthropic has launched Claude Opus 4.6—widely regarded as the two models at the very top tier right now.

    If budget isn’t a concern and you have reliable access to official subscriptions, then ChatGPT Pro and Claude Max are naturally the best choices. If you lack a suitable payment method or worry about account bans, you can also use aggregation services such as OpenRouter to access these top models on a token-based pricing scheme—though the value for money is significantly lower.

    I believe many people have recently taken advantage of Google’s promotions to subscribe to Google AI Pro or Ultra family plans. In that case, you have two options. One is to use Google Antigravity for OAuth authorization; once authenticated, you can access Gemini and Claude models with quotas refreshed periodically. From my own experience, however, this approach is prone to rate limits—especially with Claude—resulting in a subpar experience. The second option is to apply for an API key directly through Google AI Studio and pay per token via the API. Recently, Google has also given Pro and Ultra subscribers a one-time USD 300 credit plus USD 10 in monthly credits, which should last you quite a while.

    If you don’t want to spend heavily on ChatGPT or Claude models and can’t manage Google’s paid subscriptions either, it’s worth trying domestic models. To test OpenClaw, I specifically subscribed to MiniMax and Kimi’s Coding Plans. After using them for a while, I was pleasantly surprised. Their latest models—MiniMax 2.1 and Kimi K2.5—have both been recommended by OpenClaw’s developer Peter himself.

    Based on my personal experience and feedback from users on X, Kimi K2.5 has slightly stronger engineering capabilities than MiniMax M2.1, with most projects running smoothly. MiniMax M2.1, however, has its own advantages: it’s cheaper, with a top-tier plan priced at just RMB 119, compared to RMB 199 for Kimi. In terms of responsiveness, I find MiniMax M2.1 a bit faster than Kimi K2.5, making it more suitable for lighter tasks. In my own usage, Kimi Code’s RMB 99 Moderato plan was exhausted in under five days, so if you plan to use it as a main driver, you’ll likely need the Allegretto plan.

    Overall, I think the best value paid option right now is the ChatGPT Plus subscription. For USD 20 per month, you get access to GPT-5.3 Codex, a true T0-level model. OpenAI is also currently running a promotion that doubles your Codex App quota for two months.

    After comparing multiple primary models, I find GPT-5.3 Codex delivers the best overall experience, with a more balanced combination of speed and quality. It’s particularly well-suited for code audits and refactoring of legacy projects. Many other models can technically complete projects and run them successfully, but they often hide subtle issues that only surface later on.

    If you want to use another T0-level model—Claude Opus 4.6—but can’t manage a Claude subscription or are worried about account bans, you can access it via Google Antigravity (currently requires the AI Ultra plan; the Pro plan does not yet include Claude Opus 4.6). While Claude quotas are relatively limited, it’s still viable for light exploration or as one of several sub-agents handling critical but low-token tasks. If you can’t sort out subscriptions or credit purchases for overseas model services at all, then using domestic models like Kimi, MiniMax, or Qwen is also a solid option—at the very least, they’re more than sufficient for getting OpenClaw up and running and trying it out.

    If you place a high priority on personal privacy or have powerful hardware, running local models is another good choice. Through services like Ollama, you can install local large language models such as DeepSeek V3.2, qwen3-coder-next, or gemma3, then configure OpenClaw to call these local models—saving a substantial amount of money.

    Finally, if you simply want to try OpenClaw without spending extra money on AI subscriptions or credits, there are plenty of free models available. For example, NVIDIA has launched a trial program for Kimi K2.5, allowing you to apply for an API key and use it directly in OpenClaw. OpenRouter has also recently introduced a model called Pony Alpha, which can be used for free in OpenClaw—but as a trade-off, all requests and outputs are uploaded for model training purposes.

    What Can OpenClaw Do?

    With enough groundwork laid, it’s time to show some real-world use cases of OpenClaw to answer the question most people have in mind: what is OpenClaw actually good for? I’ll divide these scenarios into basic and advanced categories, so you can quickly scan according to your needs.

    Note: Most of the capabilities below are based on my own configuration and custom scripts. Results may vary depending on models, Skills, permissions, and network conditions. In addition, due to personal usage scenarios and privacy considerations, the final two advanced use cases are drawn from online sharing—please refer to them at your discretion.

    Basic

    Chatbot

    Yes—interacting with OpenClaw looks no different from chatting with a familiar chatbot. Just like what you do in the Gemini or ChatGPT web or desktop apps, you can ask questions or issue commands in OpenClaw and have a large language model respond or generate results.

    After connecting Feishu, you can chat with OpenClaw across platforms including iPhone, iPad, Android phones, Mac, and Windows. All chat records are synchronized and stored in real time. Based on your configuration, OpenClaw can also distill conversations into memories and recall them in future interactions.

    I’ve now grown accustomed to using OpenClaw as my primary way to gather information or conduct research. On one hand, it can aggregate multiple AI model services and switch between them at will; on the other, IM tools like Feishu offer a more comfortable conversational interface than most AI apps. For example, I discuss investment thoughts with the bot, look up ZIP codes, summarize documents, scrape posts from X, and more—there never seems to be a shortage of things to talk about.

    A Coding Sidekick

    The evolution of large language models by 2025 has fully proven that even people with zero programming background can easily accomplish tasks that once seemed out of reach—this is the once-viral concept of Vibe Coding. With tools like Claude Code, Codex, and Antigravity, and with Xcode now supporting ChatGPT and Claude for AI-assisted coding, ordinary users like me have gained the ability to build products firsthand.

    Imagine yourself as the product manager, with these top-tier AI models acting as full-stack engineers. All you need to do is continuously describe your ideas, and they’ll handle the coding and deliver the results.

    However, tools like Claude Code, Codex, and Antigravity still require working on a computer. Unless you remote into your machine via tools like Tailscale or Sunlogin to operate terminal-based workflows, the experience isn’t great. OpenClaw changes this completely. Wherever I am in the world, I can issue instructions in natural language through Feishu. The Feishu channel connects to the Gateway on my computer, passes the command to OpenClaw, executes various programming tasks in the terminal, and then sends the results back to Feishu.

    For example, I recently switched my primary input method on iPhone to “Cang Input Method,” but wasn’t satisfied with the default skin. I simply dropped an existing .hskin skin file to the bot and asked it to precisely adjust the skin code according to my aesthetic preferences. Once done, I imported the generated skin file directly on my phone—never touching my computer keyboard throughout the entire process.

    A Personalized News Editor

    Today, our information intake is largely controlled by big platforms and algorithms, stripping us of source control. That’s why the revival of RSS has become a hot topic. With OpenClaw, you can bypass expensive RSS subscription services and build a reading system that truly fits you.

    Step one: customize your own RSS list. You can send your frequently used RSS feeds directly to OpenClaw, ask it to discover feeds based on keywords, or—like me—simply drop an existing OPML file.

    Step two: build an RSS fetching mechanism. I had OpenClaw create a script that fetches RSS content, keeps only the text, stores it in Markdown format, retains articles from the past two days, and automatically clears older files.

    Step three: create an article filtering mechanism. If you have too many feeds or articles coming in too frequently, OpenClaw can automatically filter them based on criteria like topic, author, or online popularity—forming a screening system tailored just for you.

    Step four: set up article delivery. After filtering, I had OpenClaw create a cron task that pushes the day’s selected articles to me at a fixed time every evening, following a consistent template. OpenClaw can even summarize each article in the push, letting me quickly scan and decide which ones to read in depth.

    If you don’t like reading text, you can go a step further and have OpenClaw automatically turn the day’s content into a podcast episode or video and send it to you. It’s like running your own TV channel or media company—where you decide exactly what gets broadcast.

    A Writing Assistant

    The arrival of OpenClaw has effectively reassembled my entire writing workflow. While I’d already been using AI since last year for research, fact-checking, proofreading, and polishing, constant platform switching, model changes, and feature updates made the process fragmented.

    In OpenClaw, I created a dedicated writing group to handle all AI collaboration in my writing workflow. When writing, I place the Feishu “Writing Master” group on the left side of the screen and the iA Writer editor on the right.

    During the preparation phase, I ask Claw to gather information related to my topic, repeatedly verify its reliability, then organize it into an outline with sources attached to each point for manual verification.

    While writing, if I need to look something up, I simply ask Claw and get an immediate response—without switching tools or windows, and without breaking my writing flow.

    After finishing a draft, I let Claw handle proofreading and polishing. I created a Skill that allows me to say something like “Help me proofread/polish the article: XXX,” prompting it to read my local iA Writer library, locate the matching article, and begin reviewing it. OpenClaw checks for typos and grammatical issues, flags imprecise expressions, and provides concrete revision suggestions.

    Advanced

    Notion as an External Brain

    I originally subscribed to Notion’s Business Plan, which allowed me to operate content across my entire Notion workspace using the built-in Notion AI. The biggest advantage of Notion AI is that, at least in theory, it offers unlimited access to top-tier models such as Gemini 3 Pro, Claude Opus 4.6, Claude Sonnet 4.5, and GPT-5.2. Users who upgraded before January this year could even keep the subscription at USD 10 per month. The downside, however, is that these AI capabilities are confined strictly within Notion, which significantly limits usage scenarios.

    Later on, I discovered that OpenClaw natively supports Notion Skills. Even the free version of Notion can be controlled by an external AI agent via Connections. As a result, I canceled my Notion subscription and switched to using OpenClaw to operate my Notion content. I created a cron task in OpenClaw that, every day, selects five words from my Notion vocabulary database based on a memory algorithm and pushes them to me using a fixed template. After reviewing them, I rate my familiarity with each word, and OpenClaw records both the score and review count back into the Notion database before moving on to the next round. If I come across new words in daily life, I can simply send them to OpenClaw, which will add them to the Notion vocabulary database and automatically fill in fields such as phonetics, part of speech, definitions, mnemonic roots, and memory aids—ensuring a continuous supply of words to study.

    Controlling the Browser

    OpenClaw can act as an AI agent to control the browser on your computer, helping automate UI-level interactions.

    Before getting started, you need to install the Chrome browser extension using the terminal command openclaw browser extension install. Once installed, click the extension icon in the top-right corner of the browser and make sure it’s enabled on the current page (the icon will display “ON”). From there, you can start directing OpenClaw to work inside the browser.

    For example, I once discovered a great content creator on Xiaohongshu and wanted to scrape all of their posts for study. As a domestic social platform, Xiaohongshu obviously doesn’t provide APIs for this kind of access, nor can it be queried, posted to, or searched via Skills like X. So I had OpenClaw directly control the browser to “manually” scrape those posts. It’s slower, but at least I don’t have to constantly click the mouse myself. That said, Xiaohongshu’s anti-scraping mechanisms can be quite annoying—once you scrape too many posts, it forcibly redirects you back to the homepage.

    Of course, if you’re comfortable with it, you could also let OpenClaw help you clean up your inbox, reply to emails, or even book a flight to Paris. If you’re not comfortable with that level of access, then simply don’t install the browser extension.

    AI Phone

    Some time ago, the Doubao phone sparked a lot of discussion. Users could control the Doubao model via voice to directly operate apps on their phones—ordering a milk tea, scrolling short videos, claiming red packets, and so on. Since OpenClaw also has agent capabilities, it naturally caught the attention of curious users.

    I saw a post on X where a foreign user hacked together a USD 25 Android phone. They installed OpenClaw via Termux and were then able to use it to control the flashlight, recognize objects through the camera, read sensor data, and more.

    How the Developer Uses It

    OpenClaw’s creator, Peter Steinberger, also shared some of his personal use cases in an interview. These include adjusting mattress temperature, playing music, controlling lights, viewing camera feeds, and checking package delivery status. For specific examples, you can refer to the interview video shared in Fu Sheng’s post on X.

    OpenClaw Usage Tips

    OpenClaw is a relatively young open-source project and still maintains a high update cadence. Beyond adding new features, updates frequently focus on bug fixes—which is my roundabout way of saying that OpenClaw is not yet a mature product. You’ll inevitably run into various issues during use. Whether or not you’re an experienced developer with strong programming skills, I recommend handing all debugging tasks over to AI agents. If you don’t have to do it yourself, why would you?

    To make OpenClaw more usable and stable, I’ll share some general usage strategies here. For the actual implementation, just let an AI agent handle it. You can even copy and paste this entire section directly to an AI agent and ask it to propose solutions following these ideas, then execute them after your approval.

    Creating Groups

    Once OpenClaw is set up, you’ll initially be chatting with your bot in a private conversation. All interactions happen in this private chat, and everything is stored in memory. Over time, the bot will continuously pull related information from its memory files to respond. Eventually, those memory files become long and messy.

    To avoid this—and to keep cleaner, more focused timelines for different scenarios—I recommend creating multiple groups, channels, or topics. Add your Claw bot to each one, give it a different name and avatar, and use each group for a dedicated purpose when interacting with OpenClaw.

    By default, when chatting with Claw in a group, you need to prefix messages with @Claw (for example, if my bot is named adawinterbot, I must include @adawinterbot for OpenClaw to receive the message). Once OpenClaw receives it, it adds a 👀 emoji reaction to your message.

    If you’re tired of typing @Claw every time, have your local AI agent—or OpenClaw itself—modify the group configuration to support both @Claw mentions and direct messages.

    Also, remember to propagate configurations you’ve completed in private chats—such as model selection or Skill usage—into your group chats with the help of an AI agent.

    Reducing Token Costs and Improving Efficiency

    If you’re using paid AI subscriptions or API-based billing, you’ll inevitably feel the pain of rate limits or rapidly growing bills. This is where optimizing OpenClaw’s token usage becomes important—both to slow down token consumption and to improve output efficiency by trimming context length.

    The first method is to frequently use the /new and /compact commands. /new starts a completely fresh session with a cleared context—essentially “starting over”—which is useful when switching topics or avoiding interference from old context. /compact compresses the current session’s context, preserving key information while reducing token usage, allowing the conversation to continue more efficiently. If you want a fresh session but still retain some continuity, you can send /new followed by something like “Continue the previous task: XXX.” I’ve had OpenClaw set up a script that automatically runs /new every day at 4 a.m., which significantly reduces context buildup.

    The second method is enabling OpenClaw’s built-in QMD. This acts as a memory-retrieval middleware layer. Before each conversation turn, it selects the most relevant fragments from historical memory and injects them into the model, allowing OpenClaw to retain long-term context while controlling token costs. You can ask OpenClaw to enable QMD and tune parameters such as the maximum number of memory entries per turn, truncation length, and retrieval timeout—choosing between balanced, cost-saving, or performance-oriented presets.

    Subagents

    You may have noticed a new feature released alongside Claude Opus 4.6 called Agents Team. It enables a “main agent + multiple parallel sub-agents” architecture, where the main agent decomposes, assigns, and aggregates tasks, while sub-agents operate with relatively independent contexts to handle frontend, backend, testing, auditing, and more in parallel.

    OpenClaw also natively supports sub-agent collaboration through session orchestration, though its implementation differs from Claude’s official feature. You can ask Claw to build an advanced Subagents system based on the models you’ve configured, similar in spirit to Agents Team.

    Unlike Claude’s kernel-level sub-agent coordination, OpenClaw’s Subagents rely on custom scripts orchestrated by an overarching controller. As for how to design this orchestration, simply explain your idea to OpenClaw and let it build the system. If you have no clear plan, just ask OpenClaw to implement its own recommended approach.

    Voice Input and Output

    On mobile devices, voice has become an indispensable input and output method for interacting with AI.

    On the input side, you can have OpenClaw install the “Whisper without API” Skill and create a script that automatically transcribes incoming voice messages using this Skill. This allows you to send voice messages directly in Feishu, with OpenClaw handling transcription and executing tasks accordingly. While many IM tools already offer voice transcription, OpenAI’s Whisper supports more languages and mixed-language input, generally delivering better results.

    On the output side, you can install the edge-tts Skill, which automatically converts text responses into audio and sends them to you. Currently, edge-tts offers both male and female voices. You can configure OpenClaw to generate audio for all text responses, or only in specific scenarios.

    Backup and Recovery

    It’s easy to run into issues when modifying OpenClaw’s configuration. Once something breaks, OpenClaw may freeze or the Gateway may disconnect. If this happens while you’re traveling or away from home, recovering via IM command menus alone can be extremely difficult and stressful.

    There are currently two main approaches to address this.

    The first is self-healing via automated scripts. Start by having OpenClaw create a daily automatic backup script. Then create a heartbeat monitoring script that pings the Gateway every five minutes; if three consecutive checks fail, it automatically restarts the Gateway. Finally, add a configuration rollback script so that if repeated restarts still result in errors, OpenClaw automatically restores the last known good backup.

    The second approach is repairing via remote SSH access to your local machine. OpenClaw officially supports connecting to the Gateway via Tailscale, but I personally don’t like this option. First, Tailscale can conflict with other networking tools; second, my own programming skills are limited, so even with remote access I’d still struggle to fix bugs.

    Instead, I prefer the “AI fixes AI” approach. There are many ways to remotely control a computer—the most brute-force being desktop control software like Sunlogin or ToDesk—but that feels like overkill. Remote SSH is lighter and easier to manage.

    Beyond Tailscale, you can use Cloudflare Tunnel or VPS reverse proxies to connect to your home machine. My setup uses a VPS reverse proxy: I create a VM on GCP, connect my Mac mini to it, and then use Termius on my iPhone to SSH into the Mac mini. For a smoother experience, I recommend installing tmux on the Mac mini, so SSH sessions remain persistent instead of restarting every time.

    So when OpenClaw crashes while I’m away, I open Termius, SSH into the Mac mini, and use the Codex CLI to diagnose the issue, propose a solution, and execute the fix in one continuous flow. For specific configurations and steps, consult your own AI agent.

    Conclusion

    As mentioned at the beginning, OpenClaw’s core innovation lies more in engineering integration and usability than in breakthroughs of individual model capabilities. Still, it successfully extends the concept of AI agents—once confined to desktops—across platforms, allowing us to access AI agent capabilities on virtually any device. As an open-source project, OpenClaw has also encouraged collective creativity, giving rise to a wide range of playful experiments and productivity tools, creating strong word-of-mouth momentum.

    That said, from a cautious perspective, OpenClaw does pose certain privacy and security risks. Avoid exposing API keys online, never upload sensitive information such as financial data or home addresses, and closely monitor browser automation features—complacency is not an option.

    Of course, every technological shift and real-world deployment comes with a period of growing pains. There’s no denying that personal AI assistants are entering our lives at an accelerating pace. If you’re uneasy about a personal developer’s project like OpenClaw, waiting for companies like Apple or Google to enter the field may offer stronger privacy guarantees—providing reassurance through trust in established giants.

  • SSPAI Morning Brief: State Council Launches Investigation and Evaluation of Competition in the Food Delivery Market

    SSPAI Morning Brief: State Council Launches Investigation and Evaluation of Competition in the Food Delivery Market

    Morning Brief

    1. State Council launches investigation and evaluation of competition in the food delivery market
    2. Claude Code blocks access from third-party compatible tools
    3. Well-known open-source framework Tailwind CSS lays off a large number of employees due to the impact of AI
    4. Large-scale internet outage hits Iran
    5. Cloudflare fined by Italy, threatens to stop providing network security services for the Winter Olympics
    6. iOS 26 adoption rate lower than previous versions at the same point in time
    7. Rumors You Can Just Glance At

    State Council launches investigation and evaluation of competition in the food delivery market

    On January 9, the Office of the Anti-Monopoly and Anti-Unfair Competition Committee of the State Council announced that it would conduct an investigation and evaluation of market competition in the food delivery platform services sector.

    A person in charge of the department said that recently, problems such as aggressive subsidies, price wars, and traffic control have become prominent in the food delivery platform services industry, squeezing the real economy and intensifying “involution-style” competition within the sector, drawing strong reactions from various parts of society. This investigation and evaluation will involve on-site verification, face-to-face interviews, and questionnaire surveys to gain an in-depth understanding of competitive practices among food delivery platforms, widely solicit opinions from platform-based merchants, new forms of workers, consumers, and other stakeholders, comprehensively assess the state of market competition, organize analysis and verification, transmit regulatory pressure, and propose measures for handling the situation.

    Subsequently, Meituan and Taobao Flash Delivery both issued statements saying they would actively cooperate.

    Earlier, in February 2025, JD.com entered the food delivery business, becoming the starting point of the subsidy war in the sector. At the end of April, Alibaba announced its entry into the competition and upgraded the instant retail service “Hour Delivery” under Taobao and Tmall to “Taobao Flash Delivery,” rolling out high-profile subsidies for food delivery and further announcing a RMB 50 billion subsidy plan on July 2. In response to the competition, Meituan followed suit and joined the subsidy campaign.

    In this context, in May and July 2025, the State Administration for Market Regulation and other departments held two rounds of talks with Ele.me, Meituan, and JD.com. In September, the State Administration for Market Regulation organized the drafting of the Basic Requirements for the Management of Food Delivery Platform Services (Draft for Public Comment), which was implemented as a recommended national standard in December, setting out provisions on merchant management, fees and promotional practices, labor management, and dispute resolution.

    Claude Code blocks access from third-party compatible tools

    On January 9, Anthropic confirmed that it had deployed technical measures to prevent third-party applications—represented by popular terminal-based AI coding tools such as OpenCode—from masquerading as its official tool Claude Code to access its models, in an effort to crack down on behavior that circumvents commercial API fees.

    In response, Anthropic explained that unauthorized third-party access had caused hard-to-diagnose technical failures, leading to reduced platform stability. In addition, some xAI employees were found to be using the Cursor IDE to call Claude at scale to assist in developing their own models, violating the exclusivity clause in Claude’s terms of service that prohibits using the service to build competing products.

    However, community discussions generally believe that economic considerations are the primary motivation behind this ban. The Claude Max subscription plan is priced at USD 200 and offers a generous usage allowance. By comparison, achieving a similar level of compute through pay-as-you-go API calls would easily push monthly costs beyond USD 1,000.

    In response to the blockade, OpenCode quickly launched a USD 200-per-month OpenCode Black plan, switching to access Claude via enterprise-grade APIs.

    Well-known open-source framework Tailwind CSS lays off a large number of employees due to the impact of AI

    This week, Tailwind Labs, the company behind the well-known open-source CSS framework Tailwind CSS, confirmed that it has been forced to lay off three out of four members of its development team—75% of the entire team—due to what it described as the “brutal impact” of AI. CEO Adam Wathan revealed that although usage of Tailwind CSS is growing at a record pace, the company’s revenue has plunged by nearly 80%. Without restructuring, its funds are expected to run out within six months.

    Wathan pointed out that the widespread adoption of AI coding tools has changed developers’ working habits, leading to a 40% drop in visits to the official documentation site over the past two years. Tailwind Labs’ main business model is selling “lifetime license” Tailwind UI component libraries and templates to developers who visit the documentation. Now that developers can obtain code directly from AI, they no longer need to visit the website or purchase paid component libraries, making this business model difficult to sustain.

    Released in 2019, Tailwind CSS is one of the most popular CSS frameworks. Many AI models are particularly good at writing Tailwind code and tend to use the framework by default.

    Wathan disclosed the situation during a recent controversy in the GitHub community. Previously, some contributors proposed optimizing the documentation format to make it easier for large language models to crawl and learn from, but Wathan rejected the idea. He explained that making it easier for AI to access the content for free would make the business “even more unsustainable” and could eventually turn the project into “abandonware” due to a lack of maintainers, citing the background described above.

    The layoffs sparked widespread discussion in the tech community and reignited criticism of industry giants that use open-source projects and train AI models on open-source code without giving back. Vercel and Google AI Studio later announced that they would become sponsors of Tailwind CSS.

    Large-scale internet outage hits Iran

    According to Xinhua News Agency, around 8:00 p.m. local time on January 8, internet services in Iran’s capital, Tehran, were disrupted. NetBlocks, an international NGO that monitors global internet connectivity, said Iran is implementing nationwide network controls related to ongoing protests in multiple regions. Cloudflare’s routing data shows that Iran’s IPv4 connections dropped sharply between the 8th and 9th before partially recovering, while IPv6 connections fell to zero. (Technically, IPv6 traffic is harder to censor than IPv4.)

    The Guardian, citing expert analysis, reported that this shutdown set new records in both scale and technical measures, causing about 90% of the country’s internet traffic to vanish almost instantly. A whitelist mechanism was used during the blackout—for example, Iran’s Supreme Leader Ali Khamenei was still able to post on X.

    In addition to the internet shutdown, international long-distance calls within Iran were blocked, mobile communications services were completely paralyzed, and most regions entered a digital vacuum with no signal and no service. The Starlink satellite internet system, which had helped maintain connectivity during the 2022 protests, was also subjected to targeted signal interference this time.

    Cloudflare fined by Italy, threatens to stop providing network security services for the Winter Olympics

    Recently, Italy’s communications regulator AGCOM fined Cloudflare €14 million (about RMB 114 million) for failing to cooperate with the country’s “Piracy Shield” system in blocking pirated content. In response, Cloudflare threatened to stop providing cybersecurity services for the upcoming 2026 Milan–Cortina Winter Olympics and is considering a complete withdrawal from the Italian market.

    Under Italy’s Piracy Shield legislation, internet service providers are required to unconditionally block IP addresses and domain names suspected of infringement within 30 minutes of receiving a report from copyright holders. Since its launch in 2024, the system has been highly controversial in the tech community. Critics point out that its automated blocking mechanism lacks transparency and forces DNS resolvers to implement blocks within extremely short timeframes, making it prone to mistakenly affecting legitimate websites. For example, it previously caused large-scale internet disruptions after accidentally blocking lawful services such as Google Drive.

    Cloudflare CEO Matthew Prince sharply criticized the mechanism on X, calling it a “censorship program without judicial oversight,” and accusing it of attempting to force Cloudflare’s public DNS service (1.1.1.1) to enforce blocks globally rather than only within Italy. Prince said the company will firmly appeal the decision and plans to travel to Washington and Lausanne next week to discuss the matter with the U.S. government and the International Olympic Committee (IOC), respectively.

    As countermeasures, Prince listed four potential actions: terminating millions of dollars’ worth of pro bono cybersecurity support for the Winter Olympics, stopping the provision of free services to all users in Italy, removing all servers located in the country, and canceling plans to invest in setting up local offices. With the Winter Olympics set to open on February 6, a Cloudflare withdrawal at this point could expose the event to serious risks of cyberattacks.

    iOS 26 adoption rate lower than previous versions at the same point in time

    According to data released by web traffic analytics firm StatCounter, as of January 2026 only about 15% to 16% of active iPhones worldwide are running iOS 26, while more than 60% of devices are still on iOS 18.

    Based on these figures, the adoption rate of iOS 26 appears to be less than one quarter of that of several previous versions over the same period. For example, StatCounter’s data from January 2025 showed that about 63% of iPhones were running some version of iOS 18 roughly four months after its release. In January 2024, iOS 17 reached about 54% adoption over a similar timeframe, while iOS 16 had already surpassed 60% by January 2023.

    StatCounter’s estimates are derived from web traffic analysis, tracking operating system versions through page views across participating websites worldwide. Different methodologies, however, produce different results. For instance, TelemetryDeck reports iOS 26 at 60% adoption, with iOS 18 still at 37%.

    Well-known Apple news site MacRumors noted that in the first week of January last year, 89.3% of its visitors were using iOS 18. During the same period this year, only 25.7% of its readers were on iOS 26. As Apple has yet to release official figures, the true penetration rate of iOS 26 remains unclear, but these data suggest that user hesitation toward iOS 26 is unprecedented in recent years.

    Rumors You Can Just Glance At

    • According to social media posts, the chairman of China’s domestic Linux distribution UnionTech UOS reportedly fired a kernel engineer over his attire. The company recently issued a last-minute notice requiring employees to wear suits to the annual party. When a key engineer responsible for Linux kernel development asked in a group chat, “What if I don’t have a suit?”, the chairman allegedly dismissed him on the spot.
    • According to what some users were recently told by China Merchants Bank customer service, Visa credit cards issued by the bank can be added to Apple Pay starting January 15. Previously in mainland China, Apple Pay only supported UnionPay debit and credit cards. Apple has also updated the wording on its Apple Pay supported card list from “UnionPay credit and debit cards” to “bank cards.”
    • Wccftech claims that based on data from several Czech e-commerce websites, the 512 GB version of the Steam Machine is priced at about 19,826 Czech koruna (around RMB 6,622), while the 2 TB version is about 22,305 koruna (around RMB 7,450). However, retail prices for electronics in Europe usually include around 20% value-added tax, and third-party retailers often add extra channel markups.
    • On January 11, the team behind the recently discussed solo-living safety app “Dead Yet?” responded regarding the app’s name and the decision to start charging fees. They said they appreciate the positive suggestions from users about a new name and will carefully study and consider them. In order to keep the project healthy and sustainable, and to cover the growing costs of SMS messages, servers, and other expenses, they will introduce a paid plan priced at RMB 8. The team also welcomes interest and contact from investors, and will choose partners—institutions or individuals—who can best support the project’s growth. Finally, they called on more people to pay attention to those living alone and to offer them more care and understanding. In the “Dead Yet?” app, users must set emergency contacts and check in regularly; if they fail to check in for several consecutive days, the system will automatically send an email to the emergency contact the next day.
  • SSPAI Morning Brief: Netflix Plans to Acquire Warner Bros.’ Film and Streaming Businesses for USD 72 Billion

    SSPAI Morning Brief: Netflix Plans to Acquire Warner Bros.’ Film and Streaming Businesses for USD 72 Billion

    Morning Brief

    1. Netflix Plans to Acquire Warner Bros.’ Film Studio and HBO Max Streaming Business for USD 72 Billion
    2. “Baidu Name Directory” Found Used in Multiple Official Lists
    3. Meta Plans Major Cuts to Metaverse Budget, Shifts Spending Toward AI Devices
    4. Cloudflare Suffers Another Service Outage
    5. Russia Blocks FaceTime Citing Security Concerns
    6. Doubao AI Mobile Assistant Restricts Functions in Score-Farming, Finance, and Gaming

    Netflix Plans to Acquire Warner Bros.’ Film Studio and HBO Max Streaming Business for USD 72 Billion

    According to The Wall Street Journal, on December 5, streaming giant Netflix announced that it had reached an agreement to acquire Warner Bros.’ film studio and its HBO Max streaming business for USD 72 billion in cash and stock. Under the terms of the deal, parent company Warner Bros. Discovery will carry out a spin-off plan, separating its traditional cable TV networks — including CNN, TNT, and TBS — before selling its entertainment production and streaming division to Netflix. This marks the largest acquisition in Netflix’s history and is expected to close within 12 to 18 months.

    The transaction values Warner Bros. stock at USD 27.75 per share, with the total enterprise value of the assets involved reaching approximately USD 82.7 billion. Previously, competitor Paramount had proposed an all-cash offer of USD 30 per share to acquire the entire Warner operation, including its cable network assets. However, Warner’s board favored Netflix’s proposal because it allows shareholders to retain equity in the spun-off cable TV business while also receiving Netflix shares — resulting in a combined potential value of USD 31 to 32 per share, surpassing Paramount’s full acquisition offer.

    Through this acquisition, Netflix would gain access to one of Hollywood’s largest content libraries, including Harry Potter, Friends, the DC Extended Universe, and HBO’s premium original programming. Netflix Co-CEO Ted Sarandos described the deal as a rare opportunity to secure long-term competitive advantage and pledged to preserve Warner Bros.’ traditional theatrical distribution model post-acquisition, ensuring that select films will continue to premiere in cinemas before arriving on streaming.

    The deal, however, faces significant antitrust scrutiny. Officials from the Trump administration have already voiced concerns that the merger could excessively strengthen Netflix’s global dominance. Paramount, whose bid was rejected, also sent a letter to Warner asserting that regulatory challenges may render the deal impossible to complete. Following the announcement, U.S. markets closed with Netflix shares down nearly 3%, Warner Bros. shares up more than 6%, and Paramount — the failed bidder — plunging nearly 10%.

    “Baidu Name Directory” Found Used in Multiple Official Lists

    According to The Paper, a document titled 10,000 Common Chinese Names on Baidu Wenku has recently been linked to several suspected falsification cases. Fake names listed in the document were directly copied into official rosters across multiple domains, sparking widespread public concern over procedural integrity and review mechanisms. These names were used to “build” expert panels in government procurement announcements, fill out award lists for competitions, and even appear in a public administrative penalty notice. Among them, “Zhang Jiwei, Lin Guorui, Lin Wenshu, Lin Yanan, Jiang Yiyun” appeared so frequently that they became known as “the busiest five people on the entire internet.”

    Examples include—

    • In 2024, the Liaoning Science and Technology Enterprise Knowledge Transfer Platform and the Yingkou Science and Technology Enterprise Knowledge Transfer Platform published lists of individuals who passed the non-public senior professional title review. Several names on the lists matched those in the “Name Directory”;
    • In 2023, a graduation results announcement on the website of Guangzhou General Vocational and Technical School listed 110 students, many of whose names were identical to those in the directory, and the exam scores were fabricated;
    • In 2023, the “Luoxia Cup National Painting and Calligraphy Seal-Carving Competition,” jointly organized by Shuhua Art Network and provincial associations in Shanxi and Jiangsu, featured numerous awardees whose names came directly from the directory;
    • In 2022, in a public notice by the School of Humanities at Hangzhou Normal University for recipients of a welfare lottery public fund project, several names and their exact sequence fully matched the directory. The university released a statement on December 6 admitting staff misconduct, offering a sincere apology, and accepting public criticism;
    • In 2019, the Chengdu Recreational Environment Technology Research Institute published results for a project-funding review. Zhang Jiwei appeared as project lead, teaming up with Lin Guorui, Cai Yiyun, and Chen Zhengqian to submit a project. Following on-site and correspondence reviews by experts, the project was approved. These four individuals, along with many others on the public notice, overlapped with names in the directory.

    New cases of identical names are still continuously being uncovered and exposed.

    Meta Plans Major Cuts to Metaverse Budget, Shifts Spending Toward AI Devices

    According to Bloomberg, sources reveal that Meta CEO Mark Zuckerberg plans to drastically reduce investment in the company’s Metaverse initiatives. Meta executives are currently discussing a proposal to cut the Metaverse team’s 2026 budget by as much as 30%, with a new round of layoffs potentially beginning as early as January. A Meta spokesperson confirmed the resource reduction, noting that the freed-up funds will be redirected to fast-growing projects such as AI-powered smart glasses and other wearable devices.

    The proposed cuts stem from Meta’s annual budget planning meeting held last month at Zuckerberg’s Hawaii estate. At the meeting, Zuckerberg asked all departments to identify 10% budget reductions, though the Metaverse division was assigned far more aggressive targets. Sources explained that this is largely due to the Metaverse failing to generate the level of industry-wide competition Meta had once anticipated. The reductions are expected to focus on the VR team behind the Quest headsets and the Horizon Worlds virtual social platform — the two largest cost drivers of Meta’s Metaverse spending.

    This shift marks a significant strategic reorientation for Meta. Since the 2021 rebrand from Facebook to Meta, Reality Labs — the division overseeing the Metaverse — has accumulated losses exceeding USD 70 billion. Although Zuckerberg maintains that humanity will ultimately work and live in virtual worlds, he has rarely mentioned the Metaverse in recent public appearances or earnings calls, instead turning attention to large language models that power chatbots and to AI-integrated hardware such as Ray-Ban smart glasses.

    For years, investors have criticized the massive spending on the Metaverse for draining company resources without generating meaningful revenue. Following news of the budget cuts, Meta’s stock rose 3.4% to USD 661.53. Analysts widely believe that reducing investment in the long-unprofitable Metaverse will allow Meta to compete more effectively with its industry rivals in the generative AI space.

    Cloudflare Suffers Another Service Outage

    Internet infrastructure provider Cloudflare has confirmed that it has resolved a global service outage that occurred on Friday, December 5. The incident caused widespread delays and connection errors across the internet, affecting numerous banking and financial institution websites, video conferencing platform Zoom, and professional networking service LinkedIn, among others.

    Cloudflare reported that the outage lasted about 30 minutes. The company’s investigation found that the issue was not caused by a cyberattack but by a misconfiguration introduced during an attempt to patch a security vulnerability (CVE-2025-55182) in its React server components. The configuration change inadvertently triggered a logic error that resulted in a surge of HTTP 500 internal server errors. Cloudflare noted that such logic flaws could have been avoided in strongly typed programming languages, and the new Rust-based architecture it is rolling out has already addressed this risk.

    This marks Cloudflare’s second major outage in less than a month. As a core pillar of global internet infrastructure, Cloudflare’s services are relied on by countless companies for website acceleration and cybersecurity protection. A single point of failure at Cloudflare often produces cascading effects, knocking offline thousands of unrelated websites simultaneously.

    Cloudflare publicly apologized via social media, acknowledging that both incidents exposed weaknesses in its global configuration system, including the lack of staged rollouts and automated health checks. The company stated that it is conducting a comprehensive review to ensure network stability, and will accelerate the deployment of “fail-open” safeguards and stricter release validation procedures.

    Russia Blocks FaceTime Citing Security Concerns

    According to Reuters, on December 4, Russia’s federal communications regulator Roskomnadzor officially blocked Apple’s FaceTime video-calling service within the country. In its statement, Roskomnadzor cited law enforcement claims that FaceTime had been used “to organize and carry out terrorist attacks, recruit perpetrators, and commit fraud against Russian citizens,” though it provided no concrete evidence to support these allegations.

    As a result, users across Moscow reported being unable to use the service on Thursday; calls showed only a “user unavailable” message, and recipients could not answer even if they saw the incoming call. Apple has not yet responded. FaceTime is a core communication tool within the Apple ecosystem, known for its end-to-end encryption; the ban means Russian Apple users have now lost a major encrypted communication channel.

    This move is the latest escalation in Russia’s ongoing crackdown on Western tech platforms. Authorities have already imposed varying levels of restrictions on Google’s YouTube, Meta’s WhatsApp, and Telegram. On December 3, Roskomnadzor also blocked the U.S. gaming platform Roblox, citing the spread of extremist materials and illegal content. In August, the agency accused WhatsApp and Telegram of refusing to share anti-terrorism and anti-fraud information with law enforcement and subsequently restricted voice-call functionality on both platforms.

    Alongside restricting foreign services, Russia is pushing domestically developed alternatives in pursuit of “digital sovereignty.” This year, the government launched MAX, an official communications app intended to fill the void left by departing foreign platforms. Despite external concerns over possible data surveillance, Russian state media maintains that these measures are legitimate actions to safeguard national information security.

    Doubao AI Mobile Assistant Restricts Functions in Score-Farming, Finance, and Gaming

    According to Caixin, on December 5, Doubao announced that it would implement standardized restrictions on certain AI-driven phone operations, including—though not limited to—score-farming, reward-farming, financial app interactions, and gaming-related scenarios.

    Earlier on December 1, ByteDance’s Doubao AI model released the “Doubao Mobile Assistant” technical preview, integrated into the engineering prototype nubia M153 developed with ZTE, which was sold in limited quantities. Following the news, ZTE’s stock hit its daily limit, and the ¥3499 prototype sold out the same day. On the evening of December 3, after encountering risk-control blocks that prevented normal use of WeChat, Doubao disabled the assistant’s ability to operate WeChat.

    Doubao stated in its announcement that although the nubia M153 is merely a small-batch engineering prototype running a preview version, public attention far exceeded expectations. The concept of “AI operating your phone” sparked substantial discussion. The adjustment, Doubao says, “is a necessary step to ensure the technology develops steadily and sustainably.”

    Among the newly restricted scenarios, “score-farming and reward-farming” refers to retention-oriented incentives provided by apps that depend on real user interaction and therefore do not want AI collecting rewards. For banking and fintech applications, Doubao noted that although sensitive operations require explicit user authorization, it would still suspend AI operation of such apps out of caution. For certain gaming scenarios, Doubao said that because they involve competitive ranking, AI assistance will be temporarily disabled to maintain fairness.

    Doubao added that it will continue communicating with stakeholders to promote clearer, more predictable rules and “avoid blanket prohibitions that deny users the reasonable right to use AI.” Doubao reiterated that the AI assistant will not replace users in completing authorizations or sensitive actions; during long-running tasks, clear on-screen prompts will be displayed, and users may interrupt at any time, keeping the process fully controllable.

    Tensions among internet service companies, smartphone makers, and AI model developers are becoming increasingly visible. Recently, WeChat stopped supporting AI bookkeeping features on OPPO phones, and Bilibili no longer allows AI assistants to summarize video content. As early as 2017, when Honor was still under Huawei, its Magic smartphone featured AI functions that recommended restaurants based on WeChat and Alipay behavior — prompting Tencent to file a complaint with China’s Ministry of Industry and Information Technology, accusing Huawei of violating user privacy.

    Rumors to Take With a Grain of Salt

    • On December 4, AI search engine company Perplexity announced that global football star Cristiano Ronaldo (CR7) will invest in the company. Perplexity has also signed Ronaldo as a global brand ambassador and launched an interactive fan hub dedicated to him on its platform. Terms of the deal were not disclosed. According to recent Bloomberg estimates, Ronaldo’s net worth is roughly USD 1.4 billion. A few weeks ago, Ronaldo also met with Donald Trump, sparking speculation about potential business collaborations.
    • Bloomberg reports that Johny Srouji — widely regarded as the chief architect of Apple’s in-house chips and currently a senior vice president — has informed CEO Tim Cook that he is considering leaving the company. To retain this key talent, Apple is reportedly weighing a promotion that would make Srouji the company’s Chief Technology Officer (CTO), though such a move could trigger complex internal power-balance issues. Apple has recently been facing a wave of executive departures. According to The Wall Street Journal, LinkedIn data shows that in recent months, dozens of Apple engineers working in audio, watch design, and robotics have left for OpenAI.
    • On December 6, Elon Musk posted on X stating that he hopes to open-source all of X’s code next month — “with no reservations.”
    • According to The Wall Street Journal, sources reveal that SpaceX is preparing a new secondary stock sale, seeking a valuation as high as USD 800 billion. If completed, SpaceX would surpass OpenAI to become the most valuable private tech company in the United States. CFO Bret Johnsen has reportedly informed investors of the sale plan, and executives have indicated that the company is considering an initial public offering (IPO) in 2026. Thanks to its dominance in rocket launches and the rapid growth of its Starlink business, SpaceX continues to be highly favored by investors.
  • SSPAI Morning Brief: Google Releases Gemini 3, Cloudflare Triggers Network Outages, and More

    SSPAI Morning Brief: Google Releases Gemini 3, Cloudflare Triggers Network Outages, and More

    Morning Brief Highlights

    1. Google Releases Gemini 3
    2. Cloudflare Outage Causes Partial Network Service Failures
    3. Zigbee 4.0 Standard Announced
    4. Microsoft, NVIDIA, and Anthropic Form Strategic Partnership
    5. Apple Shares Crafting Details of Apple Watch Titanium Case
    6. Rumors You Can Just Glance At

    Google Releases Gemini 3

    On November 18, Google announced the launch of Gemini 3 via an official blog post. Gemini 3 is said to feature top-tier reasoning capabilities, with a stronger grasp of creative concepts and contextual details in user requests, enabling it to better understand user needs. The Gemini 3 Pro model is described as having “PhD-level” reasoning abilities, delivering significantly improved performance over the 2.5 Pro version across all AI benchmarks. It shows outstanding mathematical abilities and highly reliable multimodal reasoning, capable of solving complex problems across broad fields such as science and mathematics. In conversational interactions, Google says Gemini 3 Pro is smarter, more concise, and more direct, “free of clichés and flattery.” The Gemini 3 Deep Think mode goes even further, offering superior performance in reasoning and multimodal understanding.

    In terms of programming capabilities, Gemini 3 is also described as Google’s “most powerful” Vibe Coding and agent-coding model to date. It excels in zero-shot generation, handling complex prompts and instructions with ease while providing richer, more interactive visual renderings. It can execute reliable long-term planning and maintain consistent tool usage and decision-making throughout simulations spanning up to an entire year. To support this, Google also introduced the new Google Antigravity agent development platform, which provides an AI IDE experience and an agent capable of directly accessing the editor, terminal, and browser. The platform integrates three models—Gemini 3 Pro, 2.5 Computer Use, and Nano Banana (Gemini 2.5 Image)—to autonomously plan and execute complex end-to-end software development tasks, including self-verification of code.

    Starting today, Gemini 3 will first roll out as a preview of Gemini 3 Pro across Google Search’s AI mode, the Gemini mobile app, AI Studio, the Vertex AI platform, and the Google Antigravity agent development platform. The enhanced-reasoning Gemini 3 Deep Think mode is now available to safety testers and will later roll out to Google AI Ultra subscribers.Source

    Cloudflare Outage Causes Partial Network Service Failures

    On November 18, several online services—including X, OpenAI, Spotify, and Downdetector—experienced outages due to a Cloudflare service failure. After the issue occurred, Cloudflare confirmed widespread service abnormalities, including problems with its dashboard and API. Full service was restored roughly six hours later. The Cloudflare CTO apologized on social media, and in a subsequent official blog post, Cloudflare confirmed the outage was not caused by a cyberattack. Instead, it was triggered by a permissions change in the database system. This change caused Cloudflare’s bot management system to receive duplicated signature entries. The resulting file—double in size—was then distributed across all machines in the Cloudflare network, exceeding the file-size limits of the software responsible for routing traffic. Source

    Zigbee 4.0 Standard Announced

    The Connectivity Standards Alliance (CSA) has officially released the Zigbee 4.0 standard. In addition to a simplified certification process, improved security, better battery life, and enhanced batch configuration capabilities, the 4.0 standard introduces a new feature called Suzi. Suzi stands for Sub-GHz and Zigbee. Unlike traditional Zigbee devices, which mainly operate in the 2.4GHz band, devices supporting Suzi can operate in the 800MHz and 900MHz frequency bands. This provides greater signal stability and enables long-range operation without additional hardware, making it especially suitable for commercial and industrial applications. Source

    Microsoft, NVIDIA, and Anthropic Form Strategic Partnership

    On November 18, Microsoft announced that it has entered a strategic partnership with NVIDIA and Anthropic. Under the agreement, Microsoft will invest USD 5 billion in Anthropic, while NVIDIA will invest USD 10 billion. In return, Anthropic will purchase approximately USD 30 billion worth of Azure compute from Microsoft, and sign a contract with NVIDIA to acquire up to 1,000 MW of additional compute capacity. The partnership will also bring more of Anthropic’s Claude AI models into Microsoft’s Foundry business. Source

    Microsoft also announced the rollout of the new Copilot Actions feature to Windows Insider testers on the same day, requiring an update to Copilot for Windows (version 1.25112.74 or later). Users can simply describe the desired task in natural language, and Copilot will attempt to automate it—such as processing travel photos, organizing the Downloads folder, converting file formats, or extracting information from PDFs. All operations run in an auditable, isolated environment controlled by the Agent Workspace policy, independent from the user’s interactive Windows desktop. Microsoft positions Copilot Actions as an experimental Copilot Labs feature, cautioning that it may still encounter errors or difficulties when handling complex interfaces. The company recommends that users stay attentive during use to intervene or take over when necessary. The feature will initially support a limited range of tasks, with Microsoft planning to expand its capabilities as feedback is collected and models are refined. Source

    Apple Shares Crafting Details of Apple Watch Titanium Case

    On November 18, Apple published a blog post detailing the case-manufacturing process behind the Apple Watch Ultra 3 and the titanium Apple Watch Series 11. According to the post, both models feature cases produced through 3D printing using 100% aerospace-grade recycled titanium powder—a technique previously considered impossible to mass-produce. Apple states that compared with traditional subtractive machining of forged components, the additive 3D printing process used for the Ultra 3 and Series 11 reduces raw material consumption by half, saving an estimated total of more than 400 tons of titanium annually. Source

    Rumors You Can Just Glance At

    Google Play has announced its Best Apps and Games of 2025. The selections vary by region: in the U.S. store, the Best App is Focus Friend, while in Taiwan, the Best App is Adobe Firefly.

    To commemorate the 30th anniversary of the Toy Story franchise, the Steve Jobs Archive released a previously unpublished interview recorded on November 22, 1996, one year after the film’s debut. In the interview, Steve Jobs discusses how Pixar’s business model was designed to ensure that artists and engineers could benefit from copyright revenue, and how leadership and team culture played a critical role in enabling that vision. Source

    According to a report from 9to5Mac citing leaker yeux1122, the battery capacity of Apple’s foldable iPhone has been “confirmed” to exceed that of the iPhone 17 Pro Max. The unit currently in testing reportedly has a battery capacity between 5400–5800 mAh. The same source also claims that the original design for the iPhone Air included only a 3000 mAh battery—rather than the final 3149 mAh—suggesting that the foldable iPhone’s final battery capacity may end up even higher than current prototype data. Source