Pogbagram

Anthropic has confidentially filed for an IPO, ahead of OpenAI.

According to Bloomberg, Anthropic filed its IPO documents with U.S. regulators on Monday, with plans to complete the IPO as early as this fall, potentially beating its competitor OpenAI to a Wall Street listing.

Anthropic is currently valued at approximately $965 billion, and a successful IPO could be one of the largest public market debuts for an AI company. The company has previously disclosed annualized revenue of $47 billion, but its cloud computing spending and workforce size keep it in a state of high investment.

Regarding OpenAI, CEO Sam Altman downplayed the IPO competition between the two companies in an interview with CNBC yesterday, saying that it would "move forward at the appropriate time." According to a previous Bloomberg report, OpenAI is also preparing to secretly file for an IPO in the coming weeks, with the same goal of listing this fall.

OpenAI officially announces plans to build robots; Robotics division expands hiring.

OpenAI CEO Sam Altman said on X yesterday that OpenAI Robotics is hiring full-stack hardware, operations, systems, and machine learning engineers with the goal of helping teams "program and build robots that are useful to society."

This job posting confirms that OpenAI has developed the world simulation research project led by Aditya Ramesh over the past year into OpenAI Robotics.

Altman stated that AI should be able to help humans in the physical world; in the short term, OpenAI will focus on robots that can support skilled workers in building future infrastructure, while the long-term vision is "everyone having a personal robot that can perform the tasks they need."

He also stated that the project is progressing rapidly, based on the collaborative design between robotic hardware and machine learning research.

Automakers release May sales figures

Several automakers released their May 2026 delivery data yesterday:

• NIO delivered 37,705 new vehicles in May, a year-on-year increase of 62.3% and a month-on-month increase of 28.4%.
• XPeng delivered 32,158 new vehicles in May, a 4% increase compared to the previous month;
• Li Auto delivered 33,350 vehicles in May;
• Xiaomi's car deliveries exceeded 30,000 units in May.
• Seres New Energy sold 33,476 units in May; cumulative sales from January to May reached 145,108 units, a year-on-year increase of 15.14%.
• Wenjie Auto delivered 34,320 vehicles across its entire model lineup in May, a 48.2% increase month-over-month. Cumulative sales from January to May increased by 28.7% year-over-year.
• BYD Auto sold 383,453 vehicles in May, a year-on-year increase of 0.26%.
• Geely Automobile sold 237,637 vehicles in May, a year-on-year increase of 1%.
• Avita delivered 7,336 vehicles in May;
• Great Wall Motors sold 100,399 vehicles in May, a year-on-year decrease of 1.79%.
• Chery Group sold 247,823 vehicles in May, a year-on-year increase of 20.5%;
• BAIC New Energy sold 21,871 vehicles in May, a year-on-year increase of 27.75%. Xiangjie Auto sold 3,418 vehicles in May; Jihu Auto sold 17,943 vehicles in May, a year-on-year increase of 32.82%.
• Voyah Automobile delivered 13,003 vehicles in May, a year-on-year increase of 30%; cumulative deliveries from January to May 2026 are projected to reach 62,041 vehicles, a year-on-year increase of 35%.
• Yipai Technology delivered 24,830 units in May, a year-on-year increase of 42%, with a cumulative annual delivery of 109,600 units;
• GAC Haobo Aion BU sold 33,140 units in May, a year-on-year increase of 23.76%;
• SAIC Motor's vehicle sales in May reached 349,000 units, a year-on-year decrease of 4.62%.
• Dongfeng Honda sold 18,563 vehicles in May, a 20% increase compared to the previous month.
• Leapmotor delivered 81,569 vehicles in May, representing an 81% year-on-year increase.
• Jike Auto delivered 34,377 new vehicles in May, an increase of 81.8% year-on-year, setting a new record for both sales volume and price growth, marking the fourth consecutive month of double-digit growth both year-on-year and month-on-month.
• SAIC-GM-Wuling's global sales in May reached 126,087 units, with the Huajing S delivering 3,603 units in the three weeks since its launch.
• Zhiji Auto delivered 10,023 vehicles in May, with sales from January to May surging 115% year-on-year.
• HarmonyOS delivered 46,122 units in May, a 41% increase month-over-month;
• Jishi Auto delivered 2,052 vehicles in May: sales increased by 66% year-on-year, with a global target of 30,000 vehicles by 2026.

Tianya Community restarted after a three-year hiatus, and its servers were overwhelmed on the first day of operation.

Tianya Community, which had been shut down for three years, restarted at midnight yesterday with the new domain name tianya.net. However, the server was immediately overwhelmed by the influx of old users, with many users reporting that the new domain name could not be accessed and a few experiencing severe lag.

Tianya Community customer service responded that the surge in visitors caused the lag and suggested trying again later. Currently, the web version only supports browsing some posts, and old account IDs are temporarily unavailable. Interactive features are planned to be gradually restored within June, and the app is still under development.

Founded in 1999, Tianya Community once boasted over 130 million registered users and over 250 million monthly active users, earning it the title of "China's No. 1 online community." In April 2023, it ceased operations due to unpaid telecommunications data center fees, and in May of the same year, it was confirmed to be closing due to financial difficulties.

After that, the team restarted and continued to save itself. In February of this year, a recovery plan was established, and operating funds were raised through crowdfunding for "Genesis Members" (1,999 yuan/share, limited to 9,999 shares).

Doubao is rumored to launch a paid version in late June.

According to 36Kr, Doubao is expected to officially launch paid content in late June, with related features updated at the Force conference during the same period. Sources familiar with the matter stated that this timing was chosen because the PC and mobile versions still require approximately one month to complete the adaptation of basic functions and the pricing system.

Doubao had previously announced a paid version service on its App Store page, indicating that it would offer three subscription tiers on top of the free version: Standard, Enhanced, and Professional. The Standard version would cost 68 yuan per month or 688 yuan per year; the Enhanced version would cost 200 yuan per month or 2048 yuan per year; and the Professional version would cost 500 yuan per month or 5088 yuan per year.

The report also stated that Doubao will further integrate e-commerce functions to improve paid scenarios in the third quarter and drive traffic to Douyin Mall through subsidies, entering the operational phase in the fourth quarter. ByteDance will not use paid user penetration rate as a performance indicator for Doubao this year; the focus will remain on integrating AI services with Douyin e-commerce and local commercialization entry points.

NVIDIA Vera Rubin enters mass production, targeting AI factories for intelligent agents.

NVIDIA announced yesterday that its next-generation Vera Rubin platform has entered full-scale production, targeting deployments in long-context, multi-step inference, and agentic AI factories. Official information shows that Vera Rubin is a rack-mount system comprising components such as Vera CPU, Rubin GPU, NVLink 6, ConnectX-9, BlueField-4, and Spectrum-6 Ethernet.

The Vera Rubin NVL72 consists of 72 Rubin GPUs and 36 Vera CPUs. NVIDIA claims that compared to Blackwell, the Rubin platform can reduce inference token costs by up to 10 times and train MoE models with 4 times fewer GPUs.

The Rubin GPU provides 50 petaflops of NVFP4 inference power, while the Vera CPU uses 88 Olympus CPU cores, targeting host-side scheduling and data transfer in large-scale AI factories.

NVIDIA also announced its first batch of ecosystem partners. AI labs such as OpenAI, Anthropic, Meta, Mistral AI, Perplexity, Runway, and xAI will use the Rubin platform to train larger models and serve long-context, multimodal systems.

AWS, Google Cloud, Microsoft, and Oracle Cloud plan to deploy Vera Rubin instances in 2026, and Microsoft will also use Vera Rubin NVL72 in its next-generation Fairwater AI superfactory.

A fire broke out at SK Hynix's Cheongju plant, but officials say it did not affect production.

A fire broke out at SK Hynix's Cheongju plant in South Korea at approximately 10:32 AM local time on June 1st. The fire originated in the gas processing room on the sixth floor of the M15 and M15X wafer fabs in the fourth plant area of ​​Cheongju. The fire was quickly brought under control after the plant's automatic sprinkler system was activated.

A small amount of fluorine gas leaked during the fire. The factory evacuated approximately 3,600 people and sent nearby employees to the on-site medical center for examination. SK Hynix stated that the medical examination results of the personnel involved were normal, there were no casualties, and the core semiconductor production line was not affected.

The company subsequently activated environmental purification equipment to treat the residual gas and completed the cleanup work at approximately 1:38 PM local time, reopening all factory buildings. SK Hynix stated that the equipment is currently operating normally, production has not been interrupted, and the cause of the accident is still under investigation.

The first Seres x Doubao model is expected to launch this year, and may adopt both pure electric and range-extended powertrains.

According to a report by Blue Whale Auto cited by Jiemian News, Seres' new car brand, a deep collaboration with ByteDance's Volcano Engine, is working on its first model, which is expected to launch this year. Sources familiar with the matter say the vehicle is likely a crossover, somewhere between an SUV and a sedan, and is planned to offer both pure electric and range-extended powertrains.

The report states that the car is expected to be produced at the Seres Phoenix plant, which is currently undergoing production line modifications. Seres Technology's vehicles primarily collaborate with Volcano Engine on capabilities such as large-scale vehicle-to-machine interaction models, and the two companies will exchange information regularly; the intelligent driving solution is not expected to utilize Huawei's Qiankun platform.

Earlier, LatePost Auto reported that Seres' original controlling shareholder, Landian Technology, has been renamed Saidou Technology, and the largest shareholder has changed to Chongqing Shapingba State-owned Assets Supervision and Administration Commission Holding Company, with Seres' shareholding ratio dropping to 32.5%.

Shanghai and Sichuan pilot "scan-to-fly" drones.

According to CCTV News, the Central Air Traffic Management Office, together with the National Development and Reform Commission, the Ministry of Public Security, the Civil Aviation Administration of China and other departments, has designated fixed dedicated airspace within the drone control airspace of Shanghai and Sichuan Province, and promoted the "government-coordinated management, industry guidance and supervision, and user-reported flight" model, also known as "scan-to-fly".

This service targets the public in low-altitude flight scenarios and mainly provides four capabilities:

• First, it allows for one-click querying, displaying the airspace status of your location in real time, and clearly marking the flyable area;
• Secondly, one-click reporting is available. Simply scan the drone's serial number and select the flight time to complete the application, which takes only about 2 minutes.
• Third, one-click navigation accurately guides you to popular compliant flight zones;
• Fourth, it offers one-click insurance purchase, allowing users to independently select and purchase third-party liability insurance online, with immediate effect.

An official from the Central Air Traffic Management Office stated that Shanghai has opened up dedicated airspace for scenic areas, while Sichuan is exploring public low-altitude activity services through the integration of "low-altitude + cultural tourism." The official stance emphasizes that the prerequisites for promotion remain aviation safety, public safety, and national security.

Xianyu responds to AI mistakenly uploading photos of cultural relics

According to reports from CBN and Bianews, a user recently reported that their Xianyu account generated a product listing for 6,000 yuan using a photo of a "Gilded Silver Pot with Dancing Horse and Cup Design" from the Shaanxi History Museum. The product title, description, and price were all generated by the platform's AI. The user only discovered the product on their account's listing page after being contacted by a buyer.

Xianyu issued an apology yesterday, stating that after investigation, the relevant product images were identified as ordinary cultural and artistic items by AI, which then generated suggested titles and descriptions. The user has since voluntarily removed the product listings. The platform stated that it will strengthen product listing reminders and user confirmation to prevent similar misunderstandings from happening again.

Customer service explained that the dispute pointed to "Xianyu Space": photos and product information posted through the Xianyu camera are displayed in this area, and products uploaded to the space are publicly visible by default, potentially being featured on the homepage or in search results. Multiple users subsequently reported that photos of dresses, pets, and bags were also identified as products.

Unitree Robotics' IPO on the Science and Technology Innovation Board has been approved, with plans to raise 4.2 billion yuan.

According to Caixin, Unitree Robotics' IPO application on the Science and Technology Innovation Board has been approved by the Shanghai Stock Exchange Listing Committee, with plans to raise 4.2 billion yuan. The prospectus shows that Unitree's main business is quadruped robots, humanoid robots, and related components, covering consumer, industrial, and scientific research and education scenarios.

In its latest application, Unitree Robotics disclosed that the funds raised will be invested in projects such as high-performance robot R&D, intelligent manufacturing base construction, and sales and service network upgrades. The company has gained significant attention in recent years for its quadruped robot products and humanoid robot demonstrations; NVIDIA also showcased a humanoid robot reference design integrated with Unitree Robotics' platform at COMPUTEX.

Zhipu launches A-share IPO plan, aiming to raise 15 billion yuan to invest in general-purpose large-scale models.

Zhipu issued an announcement last night, stating its intention to apply for an initial public offering (IPO) of A-shares and list on the Science and Technology Innovation Board (STAR ​​Market). According to the announcement, the number of shares to be issued will account for 2% to 8% of the total share capital after the issuance, that is, no less than 9.099 million shares and no more than 38.769 million new A-shares, with a total fundraising amount of RMB 15 billion.

The funds raised will be mainly invested in three areas: 12 billion yuan for the research and development of large-scale models for general-purpose artificial intelligence, 2 billion yuan for the construction of a one-stop service platform for large-scale models MaaS, and another 1 billion yuan for supplementing working capital.

In its announcement, Zhipu stated that the purpose of this listing on the Science and Technology Innovation Board is to establish an "A+H" dual listing platform, broaden financing channels, and advance the research and commercialization of large-scale models. The company also cautioned that the issuance is not guaranteed to succeed and investors should exercise caution.

Meituan reported revenue of 91.04 billion yuan in the first quarter, but recorded a loss of 6.83 billion yuan.

Meituan disclosed its first-quarter 2026 results on the Hong Kong Stock Exchange yesterday. During the period, revenue was RMB 91.04 billion, an increase of 5.6% year-on-year; operating loss was RMB 6.47 billion, compared with an operating profit of RMB 10.57 billion in the same period last year; and the loss for the period was RMB 6.83 billion, compared with a profit of RMB 10.06 billion in the same period last year.

By business segment, core local commerce revenue was RMB 64.06 billion, basically flat year-on-year, with an operating loss of RMB 2.03 billion; new business revenue was RMB 26.98 billion, up 21.3% year-on-year, with an operating loss of RMB 2.12 billion. The company stated that intensified industry competition put pressure on profitability, but the loss from the on-demand delivery business decreased significantly compared to the previous quarter.

Meituan adjusted its revenue reporting method this quarter, disclosing merchandise sales revenue separately, mainly corresponding to self-operated categories such as Xiaoxiang Supermarket, Kuailv, and pharmaceuticals and alcoholic beverages; commission revenue and online marketing service revenue are combined into merchant service revenue.

Jensen Huang: The idea that AI is stealing jobs is nonsense.

Nvidia CEO Jensen Huang responded to the AI ​​employment controversy during his keynote speech at GTC Taipei yesterday, stating that the claim that "AI reduces jobs" is nonsense and that the number of software engineers is actually increasing.

He calls the current stage the era of "useful AI," believing that tokens have become profitable revenue units and AI is becoming a generator of GDP.

Jensen Huang used an economic example to explain the meaning behind these figures. Globally, there are approximately 30 to 40 million professional software developers, with total annual salaries of about $3 trillion, supporting an industry worth about $100 trillion worldwide.

A threefold increase in output means that these developers are generating approximately $9 trillion in productivity with $3 trillion in compensation. He concludes that when an engineer's output doubles, hiring more engineers becomes more cost-effective.

If the output curve is flat, companies will naturally reduce hiring; but when the curve is so steep, companies will only want more people.

Microsoft and Nvidia unveil the Surface Laptop Ultra, packing RTX Spark into a Windows PC.

Yesterday, Jensen Huang officially launched RTX Spark at Computex 2026. This is a Windows PC super chip designed for the era of personal AI agents, featuring 1 petaflop of AI performance, up to 128GB of unified memory, and a complete NVIDIA AI and graphics technology stack for local agents, creation, and games.

Microsoft's Surface team also released the Surface Laptop Ultra, claiming it to be the first Surface laptop to combine an NVIDIA Blackwell RTX GPU, up to 128GB of unified memory, and full CUDA support, capable of running up to 120B parameter models locally.

The device features a 15-inch mini-LED PixelSense Ultra touchscreen with a peak HDR brightness of up to 2000 nits, and retains HDMI, USB-C, USB-A, SD card, and headphone jacks.

Nvidia says ASUS, Dell, HP, Lenovo, Microsoft and MSI will launch Windows laptops and small desktops powered by RTX Spark this fall, with Acer and Gigabyte following suit.

Related reading: Nvidia flips the table, Windows finally welcomes a true AI PC.

Huawei nova 16 series launched starting at 2999 yuan.

Yesterday, Huawei officially launched the nova 16 series, including three models: nova 16, nova 16 Pro, and nova 16 Ultra, with starting prices of 2,999 yuan, 3,899 yuan, and 4,699 yuan respectively.

In terms of configuration, the nova 16 series is equipped with the Kirin 9010S processor and runs HarmonyOS 6.1. The nova 16 Pro and nova 16 Ultra are equipped with a 200-megapixel red maple imaging system, and all models come standard with a 7000mAh battery and 100W super fast charging.

Dell launches the $699 XPS 13, targeting the MacBook Neo.

According to Reuters, Dell recently released the new XPS 13, positioned as one of its most affordable XPS laptops, with a regular starting price of $699 and a back-to-school price of $599 (approximately 4,000 yuan).

This machine targets the entry-level high-end laptop market within the Windows camp, featuring a thin and light touchscreen. Information disclosed by multiple media outlets indicates that the new XPS 13 will be equipped with a 13.4-inch touchscreen, a backlit keyboard, Wi-Fi 7, and a lighter chassis, targeting users including students and young professionals.

Dell released this product after the Apple MacBook Neo, directly comparing its price to the MacBook Neo, which was released in March of this year with a starting price of $599. Dell's decision to use the XPS brand to lower the price point aims to bring the high-end design language of Windows PCs to a more affordable market.

MSI launches "True Lobster PC"

According to Wccftech, MSI has officially released the MEG Vision X2 AI+ desktop PC, which is the first to feature a holographic AI pet assistant that allows voice control of hardware performance and RGB lighting effects.

MSI's new MEG Vision X2 AI+ desktop PC features a cylindrical display interface called "AI Holostage" integrated on the front panel of the chassis, and includes a dedicated holographic AI assistant called "LuckyClaw".

This design transforms the digital companion into a visual entity that can be interacted with in real time, allowing users to complete various hardware settings without having to manually access the underlying software or control center.

In terms of specific functions, users can issue commands to LuckyClaw to automatically complete PC performance mode switching, RGB lighting system adjustments, and other system-level settings within seconds.

MSI's official statement points out that this is the "next evolution" of smart gaming PCs, marking that the application of artificial intelligence has moved beyond simple text dialogue and image generation, and has officially entered the stage of real-time system-level hardware-assisted control.

Alibaba releases Qwen 3.7-Plus, a multimodal intelligent agent model.

Alibaba officially released its latest multimodal model, Qwen3.7-Plus, early this morning. It is positioned as an intelligent agent foundation that unifies and integrates visual and language capabilities, and comprehensively expands visual and language capabilities on the basis of Qwen3.7's text capabilities.

In terms of performance, Alibaba claims that Qwen3.7-Plus ranks among the top five globally and number one in China on the Vision Arena global visual model ranking list.

In terms of plain text capabilities, the model performs strongly on the coding agent benchmarks Terminal Bench 2.0, SWE-bench series, and SciCode. On the inference benchmarks GPQA Diamond, HMMT, and IMOAnswerBench, it ranks among the top Plus-level models and is close to the Max-level model level overall.

In terms of multimodal performance, the model shows significant improvements over its predecessor in GUI agent evaluations such as ScreenSpot Pro, OSWorld-Verified, and AndroidWorld, and also has corresponding enhancements in video understanding tasks such as VideoMMMU and MLVU, as well as multiple driving scenario benchmarks.

MiniMax M3 Released, Supports 1 Million Contexts

MiniMax officially released the MiniMax M3 yesterday. The official statement claims that the M3 possesses three capabilities: coding, a 1 million token context window, and native multimodal functionality. The API supports a maximum of 1 million token context windows and guarantees at least 512K tokens.

The core architecture of M3 is MiniMax Sparse Attention. According to the official description, MSA reduces the computation cost of long contexts through sparse attention and key-value blocks. With 1 million contexts, the computation per token is about 1/20 of the previous generation model, and the prefilling stage is accelerated by more than 9 times, and the decoding stage is accelerated by more than 15 times.

In terms of model capabilities, the official benchmarks include SWE-Bench Pro 59.0%, Terminal Bench 2.1 66.0%, SWE-fficiency 34.8%, KernelBench Hard 28.8%, and MCP Atlas 74.2%.

Button 3.0 launched, enabling multi-platform Agent team building.

Kouzi 3.0 officially launched yesterday, with a full update across mobile, PC, and web platforms. The new version expands individual Agent conversations into project spaces: users can create projects and place different Agents and team members into the same task management space, allowing goals, files, process records, and outputs to be stored in the same context.

• On both the web and desktop versions, 3.0 supports adding agents, creating projects, and managing tasks;
• On the mobile device, users can continue to advance tasks or schedule agents.

The official statement also mentioned that locally running agents such as Claude Code, Codex CLI, and OpenClaw can access the Kouzi project space and process local files within the authorized scope; while cloud agents run on the cloud computers provided by Kouzi, targeting long-term online and continuous collaboration scenarios.

The new national standard for energy efficiency ratings of household refrigerators came into effect today.

GB 12021.2-2025, "Limits for Power Consumption and Energy Efficiency Grades of Household Refrigerators," approved and released by the State Administration for Market Regulation and the Standardization Administration of China, officially came into effect this month. This standard is being called the "strictest ever" energy efficiency standard for household refrigerators by the industry.

The new standard includes semiconductor-cooled household refrigerators and portable vehicle-mounted refrigeration appliances in the product scope and adjusts the categories of products that are not applicable; the energy efficiency rating remains at the 5-level framework, but the requirements for each level have been significantly raised, and the evaluation method has shifted from a single indicator to a comprehensive power consumption calculation.

Xinfei Electric Appliances, one of the core drafting units, stated that the new standard will significantly raise the market entry threshold, and it is estimated that approximately 20% of inefficient products will gradually exit the market due to non-compliance with the new standard. The new regulations will directly affect refrigerator manufacturers' subsequent product testing, energy efficiency labeling, and channel sales.

Robot hotel to be piloted on the west artificial island of the Shenzhen-Zhongshan Bridge

According to a news release from Shenzhen, Shenzhen Cultural Tourism Industry Development Co., Ltd. has reached a strategic cooperation agreement with a robotics company. The two parties will jointly build a smart hotel with full-process robot services on the west artificial island of the Shenzhen-Zhongshan Bridge, with trial operation planned for early 2027.

The project will be implemented in phases, including scenario setup, equipment deployment, system integration and debugging, and operational optimization. The hotel plans to cover seven core scenarios: guest reception, luggage guidance, room service, food and beverage delivery, comprehensive cleaning, security patrol, and interactive companionship, providing 24-hour intelligent service and standardized response.

China Mobile, together with nearly 50 partners, established the eSIM Terminal Ecosystem Alliance.

According to China Mobile's mobile phone club, China Mobile held an eSIM terminal ecosystem cooperation conference on May 29, and established the China Mobile eSIM Terminal Ecosystem Alliance with the China Telecom Terminal Industry Association, GSMA, and nearly 50 core partners in the terminal, chip, and platform industries.

The alliance established four executive committees: Capability Support, eSIM Technology, Terminal Innovation, and Industry Expansion. Simultaneously, it launched the "X-eSIM Launch" initiative, focusing on cooperation mechanisms, industry empowerment, standard implementation, and marketing services. China Mobile also released the "China Mobile New eSIM Card Technical Requirements," providing a unified technical standard for subsequent large-scale deployment.

Spider-Man: New Day Confirmed for Import

Marvel and Sony's superhero film *Spider-Man: New Day* has been confirmed for release in mainland China, with the release date yet to be determined. The film, directed by Destin Cretton and starring Tom Holland as Peter Parker, will be released globally in July 2026.

Official information indicates that Zendaya will return to play MJ, with Jacob Betalon, Sadie Zink, and Lisa Kolen-Zayas also in the cast. The trailer features the first appearance of "The Punisher," and villains such as Scorpion, Tarantula, and the Hand will also appear; Jackie Chan revealed that his stunt team will serve as the film's action director.

Charli XCX's new album will be released in July, featuring Scorsese on the cover.

Charli XCX has announced her new album, *Music, Fashion, Film*, scheduled for release this July. The album cover features director Martin Scorsese. Last year, she created a pop culture event with *Brat*, connecting album visuals, party culture, and social media buzz.

The new album title directly juxtaposes music, fashion, and film, continuing Charli XCX's approach of managing pop music as a cross-media project. Scorsese's appearance on the cover expands this release from music news to the film culture sphere.

The third season of "Euphoria" has been confirmed to be ending.

HBO has confirmed that "Euphoria" will officially end with its third season. Variety reports that the show, which ran for seven years and consisted of three seasons and 26 episodes, will not be producing a fourth season after the third season finale aired.

Euphoria premiered in 2019, created by Sam Levinson and starring Zendaya. The series revolves around the addictions, identities, families, and intimate relationships of a group of American high school students, and has become one of HBO's most recognizable teen dramas in recent years thanks to its strong visual style and use of music.

#Welcome to follow iFanr's official WeChat account: iFanr (WeChat ID: ifanr), where more exciting content will be presented to you as soon as possible.

Candysign, a sugar manufacturing company, has released the AI ​​Mini Smart Segment Mirror. As the new flagship product in the AI ​​Mini Smart Segment series, the Mirror continues the modular design of the family and adds a 0.5-inch ingBar amber status screen, a Wingie butterfly stand, and an Oxie cow stand, completing a systematic update in terms of desktop display, stand design, and dual-use experience for home and travel.

Compared to the indicator lights and fixed casing of traditional chargers, the AI ​​Mini Power Bank Mirror emphasizes a design language that is "visible, placeable, and portable": it can serve as a desktop power hub and be placed at the workstation for a long time, or it can be detached from the main body and become a portable charger.

A 0.5-inch ingBar amber display makes energy status visible.

The AI ​​Mini Power Bank Mirror features a 0.5-inch OLED self-illuminating ingBar amber screen on the front, used to display real-time power, time, and device status. The screen can display a clock with a breathing light effect during normal use, and displays key information in amber numbers when status needs to be read.

This screen uses an OLED self-emissive solution, where pixels can be completely turned off against a black background, giving the amber-colored numbers a floating appearance. An outer orange physical filter and double-sided UV treatment are used to tame blue light, transforming the glaring cold light into a warmer, softer glow.

Mirror also simulates the breathing effect of hot electrons in a vacuum tube, allowing the screen to maintain a slow, rhythmic motion when in standby mode, reducing the abruptness of traditional electronic screens. This design is also a tribute to the aesthetics of 1980s terminals.

In terms of display logic, ingBar does not attempt to pile all parameters onto the screen. Key states are presented directly on the screen, while detailed parameters are viewed on the mobile app and the full-platform control interface. Its design principle is "key information is perceived on its own, and complex parameters are presented across all platforms."

This corresponds to the display system reconstruction of CanOS 2.0: the display driver is decoupled from the hardware screen, and a three-layer architecture design of animation layer, rendering layer and driver layer is adopted, allowing the screen to expand from a single hardware function to a system capability. CanOS 2.0 can adapt to devices of different sizes and shapes, and the ingBar status bar has also become the core entry point for Mirror to establish low-interference interaction with users.

Wingie and Oxie: Stands as part of the product experience

AI Mini Mirror also launched two new stand accessories: the Wingie butterfly stand and the Oxie cow stand.

#Welcome to follow iFanr's official WeChat account: iFanr (WeChat ID: ifanr), where more exciting content will be presented to you as soon as possible.

Are you ready? Starting this June, we may witness the biggest wave of IPOs in human history.

On May 20, SpaceX filed its S-1 filing with the SEC, aiming to list on Nasdaq on June 12. Its valuation exceeds $1.8 trillion, with a fundraising target between $40 billion and $80 billion. On the same day, news broke that OpenAI was secretly preparing to file its IPO prospectus, with Goldman Sachs and Morgan Stanley jointly underwriting, aiming for a September listing. Its latest round of financing corresponds to a valuation of $852 billion.

Late last night, Anthropic officially announced that it had filed its S-1 document with the SEC. Last week, they completed a $65 billion Series H funding round, causing their valuation to surge from $38 billion in 2024 to $965 billion, surpassing OpenAI to become the world's most valuable private AI company.

Combined, these three IPOs this year could raise more funds than all venture-backed IPOs in the US since 2000. For reference, data from Renaissance Capital shows that the total amount raised in the US IPO market in the past five months this year was only $28.4 billion.

Secret IPO, we'll fix the books before showing them to you.

The original meaning of IPO is "public," but among these three giants, Anthropic and OpenAI both chose to file confidentially. SpaceX also used the same method in April, only disclosing it to the public in May.

In 2012, the JOBS Act, signed by President Obama, opened a green channel for eligible "emerging growth companies": draft prospectuses could be submitted confidentially to the SEC. All reviews, feedback, and revisions proceeded normally, but the S-1 form was not immediately made public; it was only required to be disclosed 15 days before the roadshow.

An S-1 form is like a company's most naked X-ray. SEC regulations require that the registration statement include audited financial statements, complete business and risk disclosures, management's analysis of the company's performance, details of executive compensation, and the use of proceeds. In other words, revenue, cost structure, customer information, and even how much the founders are paid will all be laid bare in this form.

For a company preparing to go public, the S-1 form is its trump card. Therefore, the secret filing channel became one of the most important institutional reforms in the US capital markets in recent decades, and the most popular star policy among companies in the JOBS Act.

There are at least three benefits to secretly submitting your cards in order to protect them.

First, the negotiations with the SEC were conducted privately and kept entirely confidential. The SEC would review the draft proposal round by round and provide feedback, to which the company would respond and make revisions. Which figures needed to be added, which statements posed legal risks—the process was repeatedly refined until both parties were satisfied before it was released to the public.

Secondly, it allows for information warfare within the fierce competition among AI vendors. As a competitor, Anthropic would never want OpenAI to see its true cost structure, and vice versa. The computing power expenditure, inference cost ratio, and reliance on major clients in S-1 are nuclear-weapon-level business intelligence for competitors. Secret submission means that the company can hold onto this information for as long as possible.

Finally, it provides an escape route. A secretly submitted draft does not constitute a formal registration statement until it is made public. The JOBS Act allows companies to terminate the process at any time during this stage, simply by not proceeding. This allows for a quiet exit when market conditions are unfavorable, avoiding headlines about a "failed IPO."

They're all good cards, but you can't look at them.

So here's the question: to maximize the benefits of a secret commit, one should remain completely silent and make a fortune without saying a word, yet Anthropic posted an announcement on its website. Why are these companies all announcing their "secret commits" and doing the opposite?

The submitted content was confidential, but the act of submitting itself was public, even high-profile, creating a carefully designed information asymmetry. Publicly declaring that a secret submission had at least several effects:

First, it stimulated valuation. Anthropic's Series H funding round and confidential filing were almost simultaneous. It was impossible for the investors not to know about their upcoming IPO. The expectation of an IPO is itself part of the pricing, which directly boosted the valuation of the private round, pushing it to a high of 965 billion, surpassing OpenAI.

The IPO news also effectively reassured clients and the team. Anthropic relies on businesses for 80% of its revenue, and major clients need to know that their finances are stable and healthy, and that they can cooperate long-term. For the team, the promised stock options now offer a glimmer of hope for realization.

Finally, there's the warm-up effect: the headline alone, "Trillion-Dollar AI Company to IPO," is enough to get the underwriting firms and institutions moving. They don't even need to look at the S-1 certificate; the roadshow hasn't even started yet, and the atmosphere is already buzzing.

In the game of capital, "disclosure" has always been a conditional, structured, and tiered disclosure. What is disclosed is not the information itself, but the hierarchy of information. Who sees it first, who can understand it, and who can act on it are all predetermined from the moment it is secretly submitted.

Not making money, or the definition of "making money" is very flexible.

Of the three most anticipated companies preparing for an IPO, SpaceX and OpenAI are not yet profitable; OpenAI, for example, is projected to generate $13 billion in revenue by 2025 but will still be operating at a loss. Anthropic is the only one that claims to be profitable.

According to an exclusive report by the WSJ, Anthropic expects to achieve revenue of $10.9 billion and operating profit of $559 million in the second quarter of 2026, compared with revenue of $4.8 billion in the previous quarter. This would be the company's first profitable quarter since its inception.

It sounds like good news, making its IPO seem like a natural progression, but a closer look at the timeline reveals many intriguing details.

Anthropic signed two new large computing power contracts this year, one from SpaceX and the other from xAI.

The first two months of these two contracts were at a discounted price, a so-called ramp-up period, with full-price billing of $1.25 billion per month starting in July. If we trace back the profit cycle, these two discounted months exactly constitute the cost basis for Q2's "profitability." Once the discounts end, computing power demand will drive up expenses, and whether profitability can be maintained is uncertain.

Moreover, this is a non-GAAP calculation. Private companies are not bound by the financial reporting standards of publicly traded companies, and their profit calculations are more flexible. Tech blogger Ed Zitron did the math: if all the computing power contracts with AWS, Google Cloud, and SpaceX were calculated at full price, the monthly cost could reach as high as $3.75 billion, which would be close to $45 billion a year—the $559 million operating profit is almost negligible in the face of this figure.

The timing of the leak is also worth noting. Anthropic's profit forecast was leaked on the same day NVIDIA released its earnings report. On one hand, the largest seller of AI infrastructure said chip demand was still rising; on the other hand, the largest buyer said it had already started making profits. These two pieces of news, overlapping, conveyed the same message to the market: the money invested in AI is not wasted. For the upcoming IPO roadshow, there couldn't be a better prelude.

There is another set of questionable figures. ARR, annualized recurring income, is the annualized estimate obtained by multiplying the current monthly income by 12, a metric that once caused Manus to stumble.

Anthropic's annualized revenue growth trajectory is unprecedented: approximately $1 billion at the end of 2024, reaching $14 billion in February 2026 (Anthropic itself disclosed in its Series G announcement: "our run-rate revenue is $14 billion"), surpassing $30 billion in April, and approaching $44 billion in May.

The four figures form a nearly vertical curve. However, CFO Krishna Rao stated in sworn testimony in court this March that the company's cumulative total revenue "exceeds $5 billion." If the ARR figures are accurate, then Q1 alone generated $4.8 billion in revenue, meaning the company has barely made any money in the past few years and relied entirely on this year's Q1 growth.

Given Claude Code's phenomenal start to the year and Codex's recent rise, the surge in Q1 revenue is understandable. However, OpenAI consistently boasts over 800 million weekly active users , a number far exceeding Anthropic's. Anthropic's valuation of $965 billion, higher than OpenAI's $852 billion, begs the question: what kind of pricing are investors assigning?

But in any case, investors don't want to wait any longer.

From 2022 to 2024, the US IPO window was practically closed for three years. The combination of interest rate hikes and market panic led to high-valuation tech companies experiencing immediate price drops after listing, a truly dismal situation. These three years also coincided with the peak of the AI ​​bubble, with paper valuations increasing several times, even tenfold.

Paper wealth remains just paper wealth. VC funds have a cycle; LPs invest money and expect returns upon maturity. Fund duration is typically 10 years, and most early-stage AI investments were completed between 2018 and 2021, meaning 2026 is already approaching the settlement period.

With the interest rate environment improving, market sentiment recovering, and the window reopening, this is almost the best time.

Whoever is in charge of the company, whoever bears the consequences.

However, Anthropic is not an "ordinary company".

It is a public interest corporation registered in Delaware, abbreviated as PBC. PBC's board of directors is legally required to balance shareholder interests with its public mission. Anthropic's public mission, enshrined in its bylaws, is "to responsibly develop and maintain advanced AI for the long-term benefit of humanity." Based on this, the company has also designed a long-term interest trust mechanism called LTBT, where independent trustees elect a portion of the board seats. This proportion will gradually increase to a majority on the board, theoretically ensuring that the security mission is not overshadowed by commercial interests.

This governance structure has never been tested in the public market, and the first challenge it will face is cost.

Once the full switch takes place next month (July), the combined monthly computing power bill from AWS, Google Cloud, and SpaceX could reach as high as $3.75 billion. According to a January report by The Information, Anthropic's inference costs were 23% higher than expected.

This discrepancy ultimately has only two possible outcomes: either it can be passed on to users, meaning price increases, reductions in free credit limits, and lower quality models for free tiers; or it can be passed on to financial statements, as the company switches from non-GAAP to GAAP accounting after going public, eliminating any room for flexible accounting and requiring the company to report gross margins to the public market every quarter. However, this means pressure on profit margins and stock price volatility.

Currently, Claude Pro costs $20 per month, and Max costs $100 to $200 per month. After its launch, these prices will be influenced by two forces: users' willingness to pay and Wall Street's expectations for gross margin. Historical experience suggests that the latter often has a stronger influence.

For paying users, the public release of S-1 is a noteworthy milestone. The prospectus discloses core data such as total revenue, detailed revenue costs (including computing power expenditures), and user scale. For the first time, analysts can use this data to calculate the computing power cost per user and the profit margin. Questions such as where the pricing ceiling is and how much room there is for price increases now have a basis for calculation.

Two paths, two products, the future

Going back to the previous question: Why does Anthropic have a much smaller user base than OpenAI, yet its valuation is higher?

These are two companies with drastically different business models. Anthropic derives 80% of its revenue from enterprise clients, and eight of Fortune's top ten clients are its customers. Claude Code contributes over $2.5 billion in annualized revenue. While its user base is small, the value of each individual user is extremely high—this is the valuation logic behind an enterprise services company.

OpenAI's logic is exactly the opposite. With 800 million weekly active users, primarily consumers, and a much lower ARPU, it follows the valuation logic of a consumer internet company.

The two valuation logics will rapidly diverge into two product paths after the IPO. Anthropic's enterprise customers account for 80% of its revenue, and its product roadmap will continue to tilt towards the needs of large clients: stronger stability and compliance capabilities, and a more conservative pace of feature releases (the restricted release of Claude Mythos reflects both the aforementioned security mission and aligns with enterprise clients' requirements for predictability). Individual users will not be abandoned, but their priority will gradually shift downwards.

The OpenAI roadmap means monthly active users are the core asset, putting greater pressure on monetization through consumer spending. Prior to Codex, they considered adult content modes, the Sora app, and advertising, and it's possible that more aggressive monetization methods will be launched soon. After all, when user volume is your biggest story, you need to give users a reason to continue paying.

The divergence between these two paths is already underway, and IPOs will accelerate this process with quarterly reports. Users choosing which AI ecosystem to embrace is essentially choosing which business model to pursue.

The S-1 will eventually be released. For investors, it's financial data; for users, it's a product roadmap. Whether subscription prices will increase, whether the free tier will be scaled back, and which way the balance between security and growth will tip—the answers are all in those hundreds of pages. It's worth taking the time to read it when it's released. Even if you're not using it for stock trading, you can still understand what might happen to the tools you use every day.

#Welcome to follow iFanr's official WeChat account: iFanr (WeChat ID: ifanr), where more exciting content will be presented to you as soon as possible.

Gurman: Apple wants to replicate the success of the Apple Watch with glasses.

Last night, Bloomberg reporter Mark Gurman revealed in the latest PowerOn newsletter that Apple is planning to enter the eyewear market "the way it entered the smartwatch market," targeting the broader traditional eyewear market rather than just competing with smart glasses manufacturers like Meta.

Apple's smart glasses, codenamed N50, were originally planned for release at the end of this year and launch early next year, but have now been delayed until the end of next year . Apple is expected to target the $200 to $500 price range, competing with traditional eyewear brands such as Essilor, Safilo Group, and Warby Parker.

Bloomberg, citing data from the World Health Organization, reported that the global eyewear market is worth approximately $200 billion annually, far exceeding the wristwatch market corresponding to the Apple Watch. The report also mentioned that Apple CEO Tim Cook is one of the strongest supporters of the project, with sources close to him stating that smart glasses are a "priority" during his tenure.

John Ternus, who will take over as CEO in September, is also deeply involved in the project, and the Vision Products Group (VPG) team responsible for research and development has been operating under his leadership for the past two years. The first generation of Apple Glasses will feature an oval camera design and will be available in a variety of colors and frame styles.

Regarding other product developments, reports indicate that the hardware for the new Apple TV set-top box and HomePod mini has been finalized for several months and is currently being used by Apple headquarters employees. Both products are awaiting simultaneous release this fall once the new Siri and Apple Intelligence features are ready.

In addition, Apple has begun early development work on iOS 28 and macOS 28 for next year, internally codenamed "Bell" and "Poppy" respectively, collectively known as "Boppy".

Gurman reiterated that iOS 27 was mainly focused on minor tweaks and improvements, while iOS 28 would be "more important than iOS 27".

Reports indicate that Samsung's custom AI chip project with OpenAI has stalled.

According to South Korean media outlet Greened, Samsung's custom AI chip development project with OpenAI has stalled due to strategic disagreements. The project, originally planned to be based on the ARM architecture, aimed to develop an NPU for inference for OpenAI, and early development had already made progress.

Samsung and OpenAI signed several cooperation agreements last year. Samsung Data Systems will collaborate with OpenAI to build AI data centers, and Samsung may also supply memory chips to OpenAI. After its custom chip project stalled, Samsung's recent investment in Anthropic has been interpreted as another AI customer strategy. South Korean media also reported that Samsung may be shifting to manufacturing AI chips for Anthropic.

Apple Music experiences global service disruption

Apple's official website "System Status" page shows that Apple Music experienced a service interruption at 23:40 Beijing time on May 29, and some users encountered intermittent connection, playback or service access abnormalities during use.

The affected regions include multiple countries and regions such as Australia, Brazil, France, India, Italy, Japan, South Korea, Spain, and the United States. Some users are still able to access and play content normally. The fault monitoring platform Downdetector received related user reports during the same period, and the number of reports subsequently declined.

According to real-time information on Apple's system status page, the service outage was resolved at 8:30 a.m. on May 30, lasting approximately 8 hours and 50 minutes.

Su Weijie, a "golden second-generation" graduate of Peking University's School of Mathematical Sciences, joins OpenAI.

Su Weijie, an alumnus of the School of Mathematical Sciences at Peking University (Class of 2007) and a professor of statistics at the Wharton School of the University of Pennsylvania, announced on the X platform that he has joined OpenAI during his sabbatical and returned to the Bay Area to participate in AI model training. He also stated that he has been promoted to full professor at the Wharton School.

Su Weijie's research interests include statistical machine learning, high-dimensional inference, large-scale multiple testing, optimization, and privacy-preserving data analysis. According to the Wharton Department of Statistics and Data Science website, he was previously an Associate Professor of Statistics and Data Science, and also an Associate Professor of Computer and Information Science.

OpenAI President Greg Brockman replied to the relevant post, "Welcome to join the team." Su Weijie graduated from the School of Mathematical Sciences at Peking University with a bachelor's degree and later pursued a PhD in Statistics at Stanford University. His research interests include statistics, machine learning optimization, and the theoretical safety of large language models.

This year, he received the COPSS Presidents' Award, one of the highest honors in the international statistics field.

Hu Yanbin personally created a fan app, launching "Yanhuo" by learning Vibe coding from scratch.

Hu Yanbin announced on Weibo yesterday that his fan community app, "Yanhuo," has officially launched. He stated that this is the first time he has personally developed an app, learning Vibe coding from scratch and completing it after repeated refinements.

The Yanhuo App is positioned as an exclusive interactive platform between Hu Yanbin and his fans, featuring daily check-ins, points levels, personalized messages, and tour tracking. The official download page shows that the iOS version is available on the App Store, while Android users can currently download the APK installation package directly from the official website.

In his Weibo post, Hu Yanbin mentioned that he hopes to use this app to "create a place that belongs to us" for fans, enabling various interactive formats such as Q&A, dynamic updates, and performance announcements. He likened the entire development process to "completing a game level" and also to the repeated polishing of music.

Hu Yanbin stated that the app is currently just the beginning, and it will continue to be iterated and updated. He also welcomed user feedback.

MiniMax initiates A-share IPO preparation process

MiniMax announced on the Hong Kong Stock Exchange yesterday that its board of directors has resolved to explore a preliminary proposal to issue RMB shares and may consult and negotiate with professional advisors, securities and regulatory authorities. The company has engaged professional advisors to provide advice on meeting the listing requirements of the Science and Technology Innovation Board and has signed a tutoring agreement.

Seres and Chongqing State-owned Assets will launch a new car brand

According to LatePost Auto, the company formerly controlled by Seres Group, Landian Technology, has been renamed Saidou Technology. The largest shareholder has changed to Shaci Zhiyuan, which is controlled by the Chongqing Shapingba State-owned Assets Supervision and Administration Commission, holding 34.5% of the shares; Seres' shareholding ratio has dropped to 32.5%.

The report states that Saidou Technology will serve as the legal entity for Seres' new automotive brand, positioned as youthful and sporty. The new brand is scheduled for launch in June and will collaborate closely with ByteDance's Volcano Engine. A dedicated sales channel will be established for both domestic and overseas markets.

Earlier this week, investors in LanDian Technology signed a capital increase agreement of approximately 6.671 billion yuan, with new shareholders including CATL, Bojun Technology, and Xingyu Shares. Sources familiar with the matter said that LanDian Technology will focus on intelligent connected new energy passenger vehicles and integrate AI technology with the automotive industry.

The annual increase in electricity consumption for computing power may exceed 100 billion kilowatt-hours.

Data from the National Energy Administration shows that by 2025, my country had built 42 intelligent computing clusters with a capacity of 10,000 kilowatts each, and the total electricity consumption of computing centers nationwide reached 170 billion kilowatt-hours, accounting for 1.6% of the total electricity consumption of the whole society. The average annual growth rate of electricity consumption for computing power at the eight major hub nodes of the national integrated computing network was about 39.5% over the past three years, which is higher than the growth rate of electricity consumption of the whole society.

The National Energy Administration predicts that during the 15th Five-Year Plan period, the annual increase in electricity consumption for computing power nationwide will be more than 100 billion kilowatt-hours, reaching about 800 billion kilowatt-hours by 2030, accounting for about 6% of the total electricity consumption of the whole society.

The city-level data center electricity consumption patterns are already quite clear. From May 1st to 27th, Shenzhen's data center industry consumed 273 million kilowatt-hours of electricity, a year-on-year increase of 20.12%; Guangming District's data center electricity consumption increased by 85.31% year-on-year. In the first four months of this year, Guangdong's total electricity consumption reached 283.797 billion kilowatt-hours, a year-on-year increase of 8.77%, of which 307 data centers consumed 3.805 billion kilowatt-hours, a year-on-year increase of 18.48%.

my country launches satellite internet technology test satellite

According to CCTV News, early yesterday morning, my country successfully launched a satellite internet technology test satellite into its predetermined orbit using a Long March 2D carrier rocket from the Xichang Satellite Launch Center.

This satellite is primarily used for experimental verification of technologies such as direct satellite connection for mobile broadband and converged space-ground networks. This mission also marks the 646th flight of the Long March series carrier rockets.

New York Fed President: AI is changing jobs, but economists' jobs remain secure.

According to Bloomberg, at the economics conference in Reykjavik, Iceland, New York Federal Reserve President John Williams gave a clear response to concerns about the impact of AI on employment, becoming the most optimistic Fed official at the conference.

Williams drew laughter from the audience with the statement that "the demand for macroeconomists remains strong," before clarifying his core judgment: the productivity gains brought about by AI will not lead to long-term structural unemployment.

History tells us that productivity can continue to increase and living standards can continue to rise without causing structural unemployment. I don't believe we will face long-term structural unemployment.

He believes that AI can accelerate data processing and preliminary analysis, but economic judgments still require identifying causal relationships, understanding the institutional context, and placing model conclusions into the policy context.

He also pointed out that the main adjustment path for the labor market is retraining and skills upgrading, rather than large-scale layoffs. He specifically mentioned that current college graduates are generally integrating AI tools into their daily work, and this generation of workers is ahead of their predecessors in terms of adaptability.

Today, those who graduate and are hired are already using AI tools every day.

Nvidia's first self-developed chip for Windows PC to be unveiled this week

According to Axios, Nvidia is expected to officially launch its first Windows PCs featuring its chips as the main processor this week. The announcements will be made at the Computex trade show and the Microsoft Build developer conference in San Francisco.

According to sources who confirmed to Axios, the first batch of PCs using Nvidia chips will include Microsoft's own-brand Surface series, as well as products from third-party manufacturers including Dell. At the same time, Microsoft is also expected to unveil new software at the event to help users more easily run AI agents on their local Windows devices to complete tasks.

Nvidia recently posted on X to tease a "new era for PCs"; Microsoft Windows head Pavan Davuluri also posted on X to hint that "new content will be released to developers at next week's Build conference, but it is not a new version of the operating system."

Analyst Carolina Milanesi (Current Strategies) stated that Nvidia's entry into the PC market is beneficial to the industry as a whole and is also expected to benefit Qualcomm, which uses a similar chip architecture. Milanesi also pointed out that Nvidia's bigger opportunities lie in the data center, with the PC business serving more as a beneficial supplement.

Tesla introduces factory-installed manual sunshade for Model Y

Tesla China has launched a factory-installed manual retractable sunshade for the Model Y, priced at 1499 yuan, aiming to solve the long-standing problem of "sun exposure on the roof" that has been complained about by car owners. It will officially go on sale this Thursday.

This product uses the same fabric as the original car headliner, paired with metallic handles, maintaining consistency with the overall interior style. The sunshade consists of an integrated frame with a roller design, which the manufacturer claims results in extremely low noise and easy installation.

The accessory is secured to the sunroof seams via front and rear pins, supports push-pull operation, and features two-stage independent adjustment. It can be installed as a single unit with the rear seats folded down. Tesla states that the product has passed whole-vehicle airbag deployment tests and multiple road tests. Currently, this accessory is only compatible with Model Y vehicles manufactured in February 2025 and later.

Samsung Galaxy Watch9 may launch a Classic version.

Android Authority discovered suspected codenames for the next-generation Galaxy Watch in Wear OS update code, including Fresh 9, Wise 9, and Project V2. The report suggests these three codenames may correspond to Galaxy Watch9, Galaxy Watch9 Classic, and Galaxy Watch Ultra 2, respectively.

Samsung hasn't always released a Classic version of the Galaxy Watch. The Galaxy Watch5 and Watch7 didn't have a Classic version, while the Watch6 and Watch8 retained the Classic line. Current reports suggest that Samsung may break this tradition.

The code also revealed new settings related to "raise-to-talk". GSMArena suggests this could mean that the Galaxy Watch9, or more Wear OS watches, will gain the raise-to-talk feature previously exclusive to the Pixel Watch.

The new watch is expected to be released alongside the Galaxy Z Flip8 and Z Fold8 series, with the release date anticipated to be July 22.

Black Shark's first AI clip-on earphones are now on sale.

Black Shark's first AI smart earphone, the "Phoenix Ear Clip Pro," went on sale yesterday, with an official price of 369 yuan.

These clip-on earphones weigh 5g per earbud, feature a titanium alloy memory steel wire C-shaped bridge, and have passed 10,000 bending tests. They are equipped with Bluetooth 6.0, support the LHDC 5.0 protocol, use a 13mm titanium-plated composite diaphragm, and offer three sound effects: gaming, music, and cinema. They also support ENC call noise reduction.

AI features include simultaneous interpretation and conversation translation. Battery life is approximately 9 hours on a single charge, with a total battery life of approximately 30 hours with the charging case; it also supports IPX5 waterproofing and dual-device switching.

Pang Donglai Stops Selling Newly Packaged Mineral Water

Pang Donglai issued a statement in advance, acknowledging a major design flaw in the new packaging of its DL Natural Mineral Water. The company announced that sales of the product would cease after yesterday's sell-out and that a redesign would be completed within approximately three months, with plans to relaunch it in September.

According to the explanation disclosed, it was discovered before the launch of Pang Donglai's new packaged mineral water that its bottle shape and appearance might be visually similar to existing products of the same category on the market.

In its statement, Pang Donglai explained that the incident was a "major design error." Based on respect for and protection of original intellectual property rights, and to avoid impacting the other party's brand market, the group's product committee decided to discontinue the sale of this batch of products. The original packaging of DL Natural Mineral Water will continue to be sold normally.

Many regions have abolished fixed time-of-use electricity pricing and shifted to spot pricing.

According to a report by Beike Finance, Guizhou, Hubei, Shaanxi, and other regions have recently been adjusting their time-of-use electricity pricing mechanisms, with fixed peak and off-peak periods beginning to loosen. Han Fang, deputy director of the Planning and Development Department of the China Electricity Council, stated that this signifies a continued shift in the electricity pricing mechanism from "plan-driven" to "market-driven."

On May 16, Guizhou Province released a draft for public comment to add a midday "off-peak electricity" period outside of winter and expand the peak-valley price difference to 120%. Beike Finance noted that with the rapid growth of installed capacity of new energy sources such as wind power and photovoltaics, the traditional method of dividing peak and valley periods based on fixed load curves is no longer suitable for the new power system.

The report points out that spot price signals are more accurate and flexible compared to fixed-time-of-use pricing. Eliminating fixed-time-of-use pricing will facilitate price transmission between the wholesale and retail markets, guiding users to adjust their electricity consumption based on real-time price changes.

Han Fang also clarified that the current policy mainly targets industrial and commercial users who directly participate in market transactions, and does not involve residential, agricultural, or public utility users; the electricity price and time-of-use policy for residential households will remain unchanged.

Shenzhen has indicated that the ride-hailing market is generally saturated.

According to a report by CBN citing the Shenzhen Municipal Transportation Bureau's April report on the operation of the ride-hailing industry, the Shenzhen ride-hailing market is generally saturated, with an average of about 13.01 orders per vehicle per day.

Shenzhen's transportation authorities remind prospective practitioners to rationally assess returns and be particularly wary of six types of risks: platforms using "work/guaranteed minimum" models to collect payments on behalf of clients; disputes arising from unauthorized intermediaries; claims of "earning over 10,000 yuan per month" and subsequent vehicle return defaults; unlicensed operation scams; high-return traps in car rental contracts; and car rental companies defaulting on insurance payments, leading to insurance lapse.

This warning directly points to the oversupply problem on local services platforms. While the number of ride-hailing drivers continues to expand, the order density per vehicle has dropped to a level that makes it difficult to support high income expectations.

The sequel to "The Movie: Minecraft" is scheduled for release on July 23 next year.

According to blogger "Hollywood Watch," the sequel to *The Minecraft Movie* has been titled *A Minecraft Movie: Squared* and is scheduled for release in North America on July 23 next year. Jason Momoa, Jack Black, Danielle Brooks, Jennifer Coolidge, and director Jared Hess will all return, with Kirsten Dunst joining the cast.

The subtitle uses "squared" as its core concept, continuing the visual language of the original blocky world. The sequel's subtitle uses a specific color scheme, leading to speculation that this may hint at the Ender Dragon's involvement in the plot.

The first episode of the third season of "The Dragon Family" will go straight into naval battle.

According to Deadline, a new clip from the third season of Family of Dragons was shown at the ATX TV Festival. Creator Ryan Condal revealed that the first episode of the new season is expected to go straight into the Battle of the Gullets, and the production scale is the largest ever for the show.

Condal explained that the third season took 314 days to film, used over 25 tons of propane, and involved a large number of actors, physical sets, and naval battle scenes. The naval battle scenes in the first episode did not rely entirely on CGI; instead, real ship sets were built, and a combination of dry and wet pools was used for filming: ship movement was simulated by mechanical structures, while some actions requiring characters to fall into the water and interact with the water surface were performed in real pools.

He summarized the core conflict of this season as a Cold War-style standoff where both sides possess "nuclear weapons," and the number, ownership, and use of the dragons will be key to the war's progress.

#Welcome to follow iFanr's official WeChat account: iFanr (WeChat ID: ifanr), where more exciting content will be presented to you as soon as possible.

The world's first intelligent agent power supply, supporting native OpenClaw integration.

The AI-powered Mini Power Bank Mirror is the world's first consumer-grade charging device to natively support the integration of AI agents such as OpenClaw. Based on the SDC architecture and MCP protocol, Mirror bridges the final gap in AI-controlled hardware, allowing AI to directly regulate charging strategies via OpenClaw without needing to open an app.

#Welcome to follow iFanr's official WeChat account: iFanr (WeChat ID: ifanr), where more exciting content will be presented to you as soon as possible.