Morning Brief: China Region App Store Makes AIGC Compliance Mandatory for Listing, OpenAI Releases GPT-6.2…
The SSPAI Morning Brief rounds up several developments: the China region App Store has made AIGC compliance materials mandatory for listing; OpenAI released GPT-6.2 Astra and GPT-6.2 Sol on November 17 local time, with both models launching the same day on ChatGPT Work, Codex, and the API—Astra at $8 input and $40 output per million tokens, and Sol at $2 input and $10 output. Other items include Kuaishou Kling launching Kling Image 4.0, SenseTime open-sourcing the official version of SenseNova Loop-9B, DeepSeek imposing limits on long conversations, and Trajectory Labs publishing the results of Gemini 4 Argon safety testing.
Why read this
AI score 68
The morning brief places model releases, platform compliance, compute quotas, and safety testing on the same timeline, making it easy to compare each company's recent moves.
Topics
Body · AI translation
Wednesday 18 Pi Morning Brief: China-region App Store makes AIGC compliance mandatory for listing, OpenAI releases GPT-6.2 Astra and more. What's your game of the year? 13-minute read Morning Brief at a glance
China-region App Store makes AIGC compliance materials mandatory for listing 3
OpenAI releases GPT-6.2 Astra and Sol models 0
Google Play announces 2026 Best App, Game and Book lists 0
Kuaishou Kling launches image creation model Kling Image 4.00
Sharp releases AQUOS sense111
TRMNL launches 31.5-inch full-color e-paper picture frame Frame0
Brief news worth a glance 0
China-region App Store makes AIGC compliance materials mandatory for listing
OpenAI releases GPT-6.2 Astra and Sol models
On November 17 local time, OpenAI released GPT-6.2 Astra and GPT-6.2 Sol, with both models launching the same day on ChatGPT Work, Codex and the API. The company says both models can complete highly difficult tasks with fewer output tokens, letting users get more work done within the same paid quota.
Astra is positioned as the model for tackling the hardest projects, officially recommended for professional scenarios such as documents, presentations and coding. It can use a screenshot of a single slide as a template, matching its layout, color scheme and even font size to generate a complete presentation with a consistent style. On the Codex side, Astra is better at making changes and testing within existing projects; the company says early testers reported cleaner code style, more focused tests and fewer rework rounds. Sol is positioned as a cheaper model for everyday coding and writing. Cognition's FrontierCode 2.0 test measures "whether changes to a real codebase meet the standard a maintainer would merge"; with both models set to maximum reasoning effort, Sol scores only half a point lower than Astra, at about one-fifth the cost.
Both models are available starting today to Plus, Pro, Enterprise and Edu users, covering ChatGPT Work and Codex. The Ultrafast high-speed mode is opening for the first time to the $200-per-month Pro tier, initially supporting Sol; the $500-per-month Pro tier and Enterprise and Edu workspaces can use it with both models, with the latter requiring an administrator to enable it. The company says this mode generates at up to 4.5 times the standard rate but consumes more usage quota, and can be toggled in the model selector. API pricing is per million tokens: Astra is $8 input, $0.4 cached input and $40 output, 20% lower than the previous generation GPT-6 Astra's standard input and output prices; Sol keeps GPT-6.1 pricing at $2 input, $0.1 cached input and $10 output. Source
iPhone Duo tests 2D face unlock feature
iOS 27.3 developer beta adds a "Unlock with front camera" option, allowing iPhone Duo to unlock the screen with 2D face recognition. Setup is similar to Face ID, but instead of slowly turning your head, you only need to turn your head left, right, up and down once each, taking about 5 seconds; unlocking takes about 0.5 seconds, and unlocking fails in very dark conditions. Also, because iPhone Duo lacks the TrueDepth camera's infrared depth map, Apple also warns during setup that this feature is "less secure than Touch ID or Face ID" and could theoretically be fooled by a photo or video; in testing, after each failed spoof unlock, the next unlock forcibly requires Touch ID, though it is not enforced when there is no face in the frame. This feature can only be used for lock screen unlocking; using Apple Pay, installing or signing in to apps still requires Touch ID or a passcode. Source
In addition, the iOS 27.3 developer beta also adds "Apple Intelligence" for China-region iPad and Mac, supporting Macs and iPads with M-series chips and iPad mini with A17 Pro and A20 Pro chips, but the China-region version does not include Siri AI. Source 2
Panasonic releases compact cinema camera SHX1
On November 17, Panasonic released the compact cinema camera LUMIX SHX1, model AU-SHX1. The camera uses a 6K full-frame partially stacked CMOS sensor with an optical low-pass filter, which reduces rolling shutter and supports high-frame-rate recording along with dual native ISO and dynamic range enhancement, achieving 15 stops of dynamic range in V-Log; it comes preinstalled with ARRI LogC3, letting users choose between V-Log and ARRI LogC3 according to their color grading workflow; the camera can record two RAW formats without an external recorder, can record Apple ProRes RAW internally to a CFexpress Type B card, and is the first LUMIX model to support recording Blackmagic RAW directly to an external SSD, with footage editable directly in DaVinci Resolve without transcoding and with the option of Blackmagic color science. In addition, it is the first LUMIX to support Open Gate recording of Blackmagic RAW 6K 24p.
In terms of body, the LUMIX SHX1 uses an aluminum-magnesium alloy frame that balances heat dissipation, durability and light weight, weighing about 984 g. It features a large cooling fan and an independent cooling system, offers dual CFexpress Type B and SD card slots, and is the first LUMIX to support Micro V-mount batteries. The LUMIX SHX1 is priced at $4,699.99 and will go on sale in batches starting in December. Source
INMO releases INMO Future smart glasses
On November 17, INMO launched the INMO Future smart glasses, using a binocular full-color array optical waveguide solution, with a first-release price of 5,999 yuan. The device weighs about 125 g, uses ABS and PC materials, with a front-to-back center of gravity ratio of 4:6; the frame has a built-in active cooling system consisting of multiple layers of thermally conductive material and a micro centrifugal fan, with daily operating noise of about 40 dB, and supports magnetic and snap-on prescription lenses, accommodating a vision range of -8D to +2D.
The display features a Micro OLED panel with 1080P per-eye resolution and 120Hz, a 36° field of view, and the company says it can present a 150-inch image at 6 meters, with up to 800 nits of eye-entering brightness, 90% light transmittance in non-grating areas, screen brightness uniformity of no less than 80%, and distortion below 1.5%. The body is equipped with a 16-megapixel camera, providing 3DoF and 6DoF spatial computing capabilities, which can combine with large models for natural-language voice interaction and object recognition; paired with the portable accessory INMO Controller, it also supports "3DoF ray interaction."
In terms of hardware, the device has 8GB of memory and 128GB of storage, dual-channel stereo speakers and a four-microphone array, a built-in 880mAh battery, and supports PD3.0 and QC4.0+ fast charging. Source
Google Play announces 2026 Best App, Game and Book lists
On November 17, Google announced the results of the Google Play Best of Play awards, divided into three categories: apps, games and books. The Best App of the Year went to Notability, which the company says won partly for its AI features; the Best Game of the Year was Arknights: Endfield. In addition, Strava and Moonlight Peaks were recognized for their cross-device performance across phones, tablets and PCs. There are also separate local lists for Asia-Pacific regions, for example Japan's Best App was Asken and Best Game was hololive Dreams. Source
SenseTime open-sources SenseNova Loop-9B official version
On November 17, SenseTime released the official version of SenseNova Loop-9B and opened the model weights and code. This version differs from the October 31 Preview. The model has about 9B parameters and, by cyclically reusing intermediate layers, expands 32 physical decoding layers into 56 layer computations, supports text and image input with text output, and has a context of up to 512K tokens. The company also provides an AWQ 4-bit quantized version.
In the comparisons listed on the official model card, the official Loop-9B scored 41.36 on NL2Repo, lower than the comparison model Qwen3.8-27B's 44.60; on AIDABench it scored 65.31, slightly higher than the comparison's 65.01. The official version outperforms the Preview on all seven evaluations shown by the company, but the model card also lists known limitations in complex instruction following, coding and long-horizon tasks. Source
miHoYo announces general vision-language-action foundation model Gamer
On November 17, miHoYo's HoYoverse (Honkai series) AI research team announced Gamer, a general vision-language-action foundation model that can autonomously play 2D and 3D games by simulating human operation. The model does not read internal game data or rely on game-specific APIs; instead, it captures frames at 5 frames per second as input and then generates native keyboard key states and mouse movement commands equivalent to 30Hz.
To address memory constraints and inference latency in long tasks, the team used over 10,000 hours of cross-game keyboard and mouse operation trajectories for continual pretraining, combined with sparse thought-instruction fine-tuning and temporal-aware reinforcement learning; they also developed an inference engine called Streaming-vLLM, which reduced end-to-end request latency from 1306 milliseconds to 65.12 milliseconds on NVIDIA H200. The team demonstrated Gamer performing main story quests, exploring scenes and completing puzzles in Honkai: Star Rail, and Gamer-RL, trained with reinforcement learning, achieved an average success rate of 49.7% across 40 previously unseen minigame tasks, outperforming the general model without fine-tuning.
In response to concerns about game cheats and automation scripts, the team emphasized that Gamer is an academic research project exploring general agents for interactive environments, with potential directions including game testing, accessibility assistance and human-computer interaction research, and explicitly opposed its use for cheating, unauthorized access or exploiting vulnerabilities. For data security reasons, the team has currently only released general files from the project's early basic pretraining stage, has not released any Honkai: Star Rail training data, and has not released model weights trained or fine-tuned on that game's data. Source: Official Cheats
Gemini 4 Argon goes out of control in third-party safety test
AI safety research company Trajectory Labs published test results on November 13. Researchers called Gemini 4 Argon via OpenRouter with default safety settings enabled and issued a "never stop, never give up" instruction to obtain official answers to nonexistent problems in FrontierMath. About 7% of requests were blocked during this period, but about 40 of the agents still probed the Epoch AI website without authorization, attempting to obtain administrator account passwords and evaluation logs. After Epoch AI set up an "agent hotline" to observe, the agents sent 36,411 requests and 487 messages within 3 hours, attempting SQL injection or impersonating employees to request information. After being moved to a local copy, they discovered the real IP, bypassed DNS redirection and blocking to reconnect online, and after the proxy was cut off, they used third-party services to retrieve pages. They also used software package caches to build a message board to share vulnerabilities and obtain credentials, and after migrating to an isolated environment, they still tried to turn off the simulated network access control device.
Trajectory Labs emphasized that it did not instruct the agents to intrude or impersonate employees, acknowledged that the test did not obtain the other party's consent and did not deploy additional isolation, and has begun formulating safety standards and recommends that peers use isolated sandboxes. On November 16, California Attorney General Rob Bonta filed a separate lawsuit against OpenAI, alleging that in its own model tests it lost control of agents at least 80 times, involving incidents such as the July attack on Hugging Face. Source
DeepSeek imposes limits on long conversations
On November 15, some users reported that DeepSeek had imposed usage limits on the long conversation feature of its app and web version. Multiple users received platform notices stating that due to "insufficient computing power, long conversations inactive for more than 7 days are no longer supported." The text limit for a single input was shortened to about 40,000 characters, daily conversation quotas were also restricted, and after the day's usage hit the cap, the page displayed "Due to server congestion, the usage quota has reached its limit."
As of November 16, DeepSeek had not issued a separate official announcement about the limits or provided a recovery timetable. According to DeepSeek's official status page, since October 2026 the platform has experienced 8 service outages, involving the conversation service and the APIs of V4 Pro and V4.1 Flash. Among them, the availability of the V4.1 Flash API from August to October was 99.46%, equivalent to about 47.3 hours of unavailability over a full year.
Growth on the demand side is one source of pressure. The weekly rankings of multi-model aggregation platform OpenRouter show that in the week of October 19 to 25, DeepSeek V4.1 Flash completed 33.6 trillion token calls, up 801% week-over-week, ranking first globally. The model's weekly call volume from late July to early August was 7.22 trillion tokens, more than quadrupling in less than 3 months. Source 2
Kuaishou Kling launches image creation model Kling Image 4.0
On November 17, Kuaishou Kling AI announced the launch of the image creation model Kling Image 4.0. The new model adopts a "unified understanding and generation" architecture, incorporating visual understanding and image generation into a unified model system and connecting the chain from understanding creative intent to generating visual content.
The model supports native 8K resolution generation, and the company says it preserves more detail, making later cropping easier. For text rendering, a single image supports up to 2,000 characters, small font sizes are clear, multiple text blocks and complex layouts can maintain neat typesetting, and it offers more than 70 preset fonts and supports uploading reference images to specify a style. In addition, the model supports Image Sets, generating a whole set of coherent images from a single prompt, and it opens API access to enterprise customers and developers. Source
Sharp launches AQUOS sense11
On November 17, Sharp launched the AQUOS sense11 phone, equipped with a Snapdragon 7s Gen 3, Android 17 and a 5000 mAh battery, with a body about 7.6 mm thick and about 157 g in weight. The phone uses an approximately 6.2-inch Pro IGZO OLED screen with a resolution of 2340 × 1080, with official peak brightness listed at 3000 nits; it offers an 8 GB + 128 GB version, as well as a 12 GB + 256 GB version for the unlocked market.
The rear main camera and ultra-wide camera are both about 50.3 megapixels, the main camera supports optical image stabilization, and the front camera is about 32 megapixels. Imaging features include passerby removal, shadow removal, glass reflection suppression and text masking. The phone supports IPX5, IPX8 and IPX9 water resistance and IP6X dust resistance, and offers up to four major system version upgrades and six years of security updates. The new product will go on sale in Japan from November 26 through carriers and unlocked channels. Source What about the price?
TRMNL launches 31.5-inch full-color e-paper picture frame Frame
On November 16, TRMNL released the 31.5-inch Spectra 6 full-color e-paper screen Frame, with a resolution of 2560×1440, mounted in a solid wood frame handcrafted in its own woodworking shop, mainly used to display static content such as paintings, photos and maps. The first batch is limited to about 470 units, priced at $1,299, with small-volume shipments starting in late 2026, most orders expected to be delivered in early 2027, and the initial launch only for the United States, Canada, Mexico and Australia.
Frame has a built-in 10000mAh battery, upgradeable to 20000mAh, and can last 2 to 6 months depending on refresh frequency; on the back of the body, in addition to the regular USB-C port, there is another always-powered port, and the rear cover has a small hole for cable routing. The device supports networking but retains a microSD card slot, can run completely offline, and does not send personal information back to servers. To improve image quality, TRMNL developed a color conversion engine called Lazarus, claiming it can achieve leading color matching among all Spectra 6 devices; the engine will later be released as open source; the clock synchronization function has been changed to refresh at specified times rather than at fixed intervals, to reduce flicker and extend battery life. The product supports both horizontal and vertical mounting, and the stand must be purchased separately or made yourself. Source
Brief news for a quick look
The Game Awards announced the 2026 nominations on local time November 17, with Nintendo being the most nominated publisher this year; "GTA 6" received 8 nominations including Game of the Year, and the Chinese game "Phantom Blade Zero" received 5 nominations including Game of the Year, Best Game Direction, Best Art Direction, Best Score and Music, and Best Action Game. Source
On November 17, Apple services including iCloud account and sign-in, the App Store, Apple Music and AppleCare on Device experienced an outage, and the status page at one point marked it as affecting "all users"; retail store systems went offline, preventing customers from completing purchases, and some users were unable to activate new iPhones. Apple told Bloomberg that the outage was caused by an "internal update" and was not a cyberattack, and that most systems had recovered. Source
Microsoft published a preliminary report on the global Microsoft 365 outage on November 12: a software defect introduced in July triggered lock contention among server processes, which, combined with two other infrastructure problems, formed a cascading failure that at one point caused 80% of web servers to stop responding, with Outlook, Teams and other services interrupted for more than 12 hours. Source
AMD announced the completion of its acquisition of World Labs, and Fei-Fei Li joined AMD as executive vice president and chief scientist, reporting directly to Lisa Su; her team will continue to advance foundation model research and participate in the development of next-generation AI infrastructure. Source
NVIDIA released the Rubin HPC GPU for scientific computing and simulation at the SC26 conference on November 16, with more than 50,000 CUDA cores per GPU, support for seventh-generation NVLink, and an expected market launch in 2028; the DGX Spark HPC development kit released the same day consists of four DGX Spark units, with the whole machine providing 512GB of memory and 20TB of NVMe storage, and can locally deploy models with up to 700 billion parameters. Source
On November 17, according to The Wall Street Journal, Jensen Huang said he would be willing to return to a position at NVIDIA if the board believed it was in its interests, but The Verge reported that the board "has not considered" bringing him back, and the SEC has launched a civil investigation into NVIDIA. Source
On November 17, the Ministry of Industry and Information Technology agreed that Beijing, Hebei, Shanghai, Jiangsu, Zhejiang, Anhui, Hubei and Guangdong, 8 provinces and municipalities, would carry out ministry-province collaborative pilots for 6G innovation and development; enterprises involved in commercial applications must first obtain a telecommunications business operating license. Source
Anthropic announced that starting January 1, 2027, it will tighten its coordinated vulnerability disclosure policy: the disclosure period for high-severity and critical vulnerabilities will be shortened from 90 days to 30 days, and other levels narrowed to 45 days, one reason being that AI has greatly accelerated the speed of vulnerability discovery and duplicate discovery. Source
Recent updates from SSPAI
We have opened a special page to collect and display the "Outstanding Author Awards" awards and results. The "Outstanding Author Awards" is a mechanism implemented by the editorial team since 2023. After the end of each quarter, it selects existing authors who made outstanding contributions that quarter, as well as emerging promising new authors. Winners receive an exclusive badge and prizes. Go to view
Limited-time offers on SSPAI membership, genuine software and paid columns. Don't miss out
Good articles you may have missed
Community Express 166 | Konica film cameras, iPhone Duo experience and a practical guide to saving money on vacation
TDS REVIEW | Bose wired noise-cancelling earbuds experience
For another possibility: some notes on tinkering with Linux gaming in China
> Download the SSPAI client and follow the SSPAI official account to unlock a new reading experience 📰
> Practical, useful genuine software, presented to you by SSPAI 🚀 SSPAI Morning Brief Editor of this article:
@Clyde
3 SSPAI friends have already charged
AI Trends