Today's Overview
OpenAI announces full API access for GPT-5 with significant price cuts; Kimi Code caught in "double standards ban" controversy, differential treatment between domestic and foreign users sparks strong dissatisfaction in the developer community; Meta launches paid subscriptions for Instagram/Facebook/WhatsApp, AI features become a new selling point.
Domestic Hot Topics
🔥 Kimi Code caught in double standards ban controversy: overseas instant unban, domestic ignored
Kimi Code, a subsidiary of Moonshot AI (月之暗面), implemented expanded risk controls to crack down on scalping and ticket reselling, which mistakenly banned numerous normal domestic paid users. However, for similar feedback from overseas users, the company responded quickly and batch-unbanned them, sparking "differential treatment" controversy. After open-source developer Leechael announced suspension of updates in protest, the company publicly apologized and promised improvements on May 26th.
Source: 36Kr/LeiTech AGI | Details: https://36kr.com/
🔥 Tencent Research Institute: AI commercialization at a crossroads, why can't advertising and subscriptions work alone?
Tencent Research Institute published an article exploring the challenges of AI commercialization, pointing out that neither the advertising model nor the subscription model alone can sustain long-term profitability for AI products, with AI Agents potentially being the key to breaking through.
Source: 36Kr/Tencent Research Institute | Details: https://36kr.com/
🔥 Robotics sector continues to heat up: 27 billion yuan achieved in just 2 years of entrepreneurship
Chinese robotics companies like Unitree are rapidly rising, with core component suppliers like reducers becoming the most lucrative "shovel sellers," and investment enthusiasm across the industry chain continues to surge.
Source: 36Kr/Pencil Note | Details: https://36kr.com/
🔥 Short drama + AI propels Chinese Online: 1 billion yuan loss can't stop 1300% stock surge
The "AI Short Drama Factory" concept continues to gain traction. Chinese Online, leveraging AI generation technology, is expected to achieve an annualized production capacity of 3,000 short dramas, with the capital market seeing astonishing growth.
Source: 36Kr/Mirror Entertainment | Details: https://36kr.com/
🔥 Card drawer with 7,000 yuan monthly salary: rapid rise and fall of new positions in the AI short drama industry
As AI short dramas become the new trend, "card selection masters" (a new role responsible for selecting and arranging AI-generated images) have emerged, with monthly salaries around 7,000 yuan, but the industry has low barriers and high personnel turnover, showing characteristics of rapid birth and death.
Source: 36Kr/豹变 | Details: https://36kr.com/
🔥 Yushu Supplier Becomes the Biggest Winner: Reducer Giant Takes the Cake
Yushu Technology has not yet gone public, but its Ningbo supplier (reducer manufacturer) has already achieved performance explosion, becoming an invisible champion in the robot boom.
Source: 36kr/源媒汇 | Details: https://36kr.com/
🔥 The Biggest Cognitive Bias in 2026: Countless People Miss the AI Investment Opportunity
Market analysis points out that many investors exhibit cognitive bias in AI-related assets. The core reason for missing this round of AI market lies in underestimating the speed of technology penetration.
Source: 36Kr/格隆汇 | Details: https://36kr.com/
🔥 Can Afford Tokens, But Can You Find "Reliable"? AI Tool Quality Variance Raises Concerns
As the cost of large model APIs continues to decline, "affording tokens" is no longer the issue, but the actual effectiveness and reliability of AI tools vary greatly, and users are beginning to focus on the balance between cost-effectiveness and usability.
Source: 36Kr | Details: https://36kr.com/
International Hot Topics
🔥 OpenAI Fully Opens GPT-5 API with Significant Price Reduction
OpenAI announces that GPT-5 is now fully available via API. Developers can call GPT-5 for application development, with pricing strategy significantly optimized compared to previous products, further promoting AI application adoption.
Source: Hacker News | Details: https://news.ycombinator.com/
🔥 Meta Launches Instagram/Facebook/WhatsApp Paid Subscriptions, AI Features Become New Selling Point
Meta officially launches the Meta One subscription plan globally, covering Instagram, Facebook, and WhatsApp, adding AI personalization features and creator tools, with AI becoming the core differentiator for paid value‑added services.
Source: TechCrunch | Details: https://techcrunch.com/2026/05/27/meta-officially-launches-instagram-facebook-and-whatsapp-subscriptions-with-more-to-come-including-ai-plans/
🔥 AI Coding Unicorn Cognition Raises $1 Billion, Valuation Reaches $25 Billion
Devin’s AI programming company Cognition completed a $1 billion financing round at a pre‑money valuation of $25 billion, with annualized revenue reaching $492 million, and its valuation more than doubled in eight months.
Source: TechCrunch | Details: https://techcrunch.com/2026/05/27/ai-coding-startup-cognition-raises-1b-at-25b-pre-money-valuation/
🔥 Robinhood Opens AI Agent for Autonomous Stock Trading
Robinhood announces that users can create independent trading accounts for AI Agents and fund them, allowing AI agents to autonomously execute stock trades and credit‑card spending decisions, marking a concrete step in the integration of finance and AI Agents.
Source: The Verge | Details: https://www.theverge.com/ai-artificial-intelligence/938095/robinhood-ai-agent-stock-trading
🔥 YouTube Strengthens AI Content Labels: Will Be Placed Where Users Can Actually See Them
After Google I/O, YouTube announced a redesign of its AI-generated content disclosure mechanism, moving AI labels to more prominent positions such as Shorts to improve content transparency.
Source: The Verge | Details: https://www.theverge.com/streaming/937915/youtube-ai-labels-shorts-automatic-identification-updates
🔥 Pope Leo XIV Issues AI Encyclical, Warning That AI "Use Is Never Pure"
Pope Leo XIV issued an encyclical titled "Magnifica Humanitas" discussing AI's impact on society, warning that "the use of AI is never something purely good"; analysts note that some passages in the encyclical appear to have been written by AI.
Source: The Verge | Details: https://www.theverge.com/ai-artificial-intelligence/937933/pope-ai-encyclical-tech-industry-reactions
🔥 AI Recruitment Algorithms Have Higher Rejection Rates for Black and Asian Job Seekers
The Register reports that research found mainstream AI recruitment tools have higher rejection rates for Black and Asian job seekers, raising systemic concerns about algorithmic discrimination.
Source: Hacker News | Details: https://www.theregister.com/ai-ml/2026/05/27/ai-hiring-algorithms-reject-black-asian-job-seekers-at-higher-rates/5247387
🔥 ElevenLabs New Music Generation Model Supports Mid-Stream Style Switching
ElevenLabs released a new generation music generation model that can regenerate a specific section mid-play without affecting other parts, enabling more refined control over AI music creation.
Source: TechCrunch | Details: https://techcrunch.com/2026/05/27/elevenlabss-new-music-generation-model-can-switch-genres-mid-track/
🔥 ArXiv New Papers: AI Agent Long-Term Memory Becomes Research Hotspot
Several latest ArXiv papers focus on AI Agent memory mechanisms, proposing a rethinking of "whether Agent memory should be treated as a database," as well as systems like "ScientistOne" for scientific research automation.
Source: ArXiv | Details: https://arxiv.org/rss/cs.AI
Comments