AI News

AI Trending Rankings

The 10 most-discussed topics in AI over the past 48 hoursUpdated 11/16 18:00 (UTC+8) · ranked by discussion activity

Updated 11/16 18:05 (UTC+8) · refreshed hourly

NO.01↑ 84%Breakout

Alibaba Releases Qwen4-Max, with the Preview Available First on Multiple Products

On November 16, the Qwen team officially released its flagship model, Qwen4-Max, saying it is based on the new Qwen4 architecture and uses a sparse MoE with 2.4 trillion parameters, activating 71B per token. It features native vision capabilities and a 1-million-token context window. Earlier, on November 11, the Qwen4-Max preview launched on Alibaba Token Plan (with a limited-time offer as low as 10% of the regular price) and became available on products including the Qwen App, Qwen Office, Qoder, and QwenStudio. The Qwen Go! membership in the Qwen App has been offered at a limited-time discounted price starting at 9.9 yuan since November 11. Following the official launch, the Qwen App said members could try the new model for free during its launch period, with Qwen4-Max usage not counting against membership quotas. See the product page for the promotion period. The Qwen team says Qwen4-Max is more efficient in token usage and cost than Qwen3.8-Max: internal evaluations show an average cost per task reduction of more than 60%. In the Work360 evaluation, token usage fell by about 24% (Xhigh) and 35% (High), while average cost per task dropped from $4.43 to $1.67. Its Work360 benchmark score is 52.69, higher than …

Latest developmentsQwen Releases Its Next-Generation Flagship Model, Qwen4-Max

+328 sources · 37 participants321Popularity index
NO.02Unchanged

NVIDIA employees and investors pressure the board, which seeks another successor

After NVIDIA’s board ousted Jensen Huang, employees and investors are applying pressure from two directions: According to an internal open letter forwarded by X.PIN, hundreds of employees signed the letter, demanding that the board resign and that Jensen Huang be reinstated as CEO. The letter accuses the board of removing Huang without specific evidence and failing to negotiate in good faith, and says that if the demands are not met, employees may move to SpaceX and Terafab, where Elon Musk has promised to arrange positions; these accusations and claims about employees’ destinations after leaving are statements made in the letter. Investors Brad Gerstner and Gavin Baker also wrote to the board, urging it not to rush to replace Huang and to preserve the possibility of his return as CEO. Meanwhile, according to the FT, the board has contacted Arm CEO Rene Haas as a possible successor to Huang. Haas has led the SoftBank-owned British chip design company since 2022 and worked at NVIDIA for about seven years more than a decade ago, but he is only one of several candidates and may not be offered or accept the position. The Decoder, citing the WSJ, reported that Huang’s removal stemmed more from the high-risk businesses he led than from rumors about his private life: NVIDIA provided tens of billions of dollars in financial guarantees to AI cloud providers and data center developers in exchange for a share of revenue, with the largest deal involving…

Latest developmentsInvestors write to Nvidia’s board urging it to preserve the option of Jensen Huang’s return as CEO

+181 sources · 23 participants172
NO.03New entryBreakoutNew

Tencent Hunyuan Releases Hy4 flash, Now Available on WorkBuddy and OpenCode

Tencent Hunyuan releases the Hy4 flash model, with 224B total parameters (including 64B Engram parameters) and 12B activated parameters per token. It supports multimodal input and a 1-million-token context window. It is positioned as the lowest-priced Hy model for everyday productivity tasks. API pricing is 0.5 yuan per million input tokens and 2 yuan per million output tokens. Tencent Hunyuan says that in blind evaluations of everyday tasks, Hy4 flash performs comparably to Claude Haiku 5.5 (high) at one-eighteenth the task cost. Tencent WorkBuddy has integrated Hy4 flash into its platform, citing faster responses, lower cost per task, and built-in visual understanding. The platform says it is available for a limited-time free trial, with a daily free quota; credits will be required afterward. OpenCode announced that Hy4 flash will be free with near-unlimited usage for the next two weeks, and that this usage will not count against users' Go plan quotas. Tencent Hunyuan lists USD API pricing at $0.003 per 1M cached tokens, $0.070 per 1M input tokens, and $0.281 per 1M output tokens. IT Home reports that the cost per task is about 5% of Hy4's.

Latest developmentsHy4 flash Available for Free on OpenCode Go

+106 sources · 15 participants139

Continue to No.04–10

  1. 04+21139↓ 18%
  2. 05
    Apple and OpenEvidence launch a study on the use of health AI

    Apple launched the Apple Health AI Usage Study in the United States and partnered with healthcare AI company OpenEvidence to release a research app studying how people interact with conversational health AI tools. The Verge reported that the study also aims to evaluate the accuracy, detail, and relevance of these responses. The study is open to people aged 21 and over in the United States, with a goal of recruiting up to 200,000 participants over more than 12 months. Participants must have an iPhone that supports Apple Intelligence and run iOS 26 or later; Apple said an iPhone 15 Pro or newer is required, while TechCrunch reported that an iPhone 15 or later is sufficient. Apple said that data will be anonymized before being sent to third-party cloud service providers and will not be used to train any AI models.

    +577↑ 29%
  3. 06
    Anthropic releases Fable 5.2 and Mythos 5.2, topping multiple evaluations

    Anthropic released Claude Fable 5.2 and Claude Mythos 5.2, both based on the same underlying model. Fable 5.2 is available on all paid plans and includes safeguards that block high-risk cybersecurity and biological research tasks; Mythos 5.2 removes some restrictions and is available only to organizations that apply through the cybersecurity or life sciences verification programs. Fable 5.2 is priced at $8 per million input tokens and $40 per million output tokens, down 20% from Fable 5.1’s $10/$50 pricing, supports a 1-million-token context window, and is now available on platforms including Claude Code, OpenRouter, Cursor, Perplexity, Google Cloud, and OpenClaw. In evaluations, it topped the Artificial Analysis Intelligence Index v5 with a score of 65; ranked first in Arena Code Arena: WebDev with 1832 points; led the Remote Labor Index with 41.3%; and ranked first in AA-Robotics v0.1 with 75 points, ahead of GPT-6 Astra (69) and GPT-6.1 Sol (65), and became…

    +875↓ 16%
  4. 07
    Musk and others sign petition calling on NVIDIA to reinstate Jensen Huang

    A petition calling on NVIDIA to reinstate ousted CEO Jensen Huang had received more than 115,000 signatures as of Saturday afternoon. According to the petition website, signatories include 5,965 current NVIDIA employees, 1,108 former employees, and 63,063 shareholders. Technology executives including Tesla and SpaceX CEO Elon Musk, Replit CEO Amjad Masad, and Epic Games CEO Tim Sweeney have signed the petition. The petitioners are calling on NVIDIA to restore Huang to his position as CEO.

    +458Unchanged
  5. 08
    OpenAI disbands robotics team, with researchers moving to other departments

    According to Fortune, OpenAI is disbanding its robotics team and transferring robotics researchers to the main research team led by Mark Chen, while the applied robotics and hardware teams will join Greg Brockman’s Applied organization. The Decoder said that OpenAI is not abandoning robotics, but instead hopes to integrate the researchers into other teams and replace specialized robotics AI with general-purpose models to accelerate progress. This account comes from the reporting outlet; OpenAI has not yet announced specific plans or results for the reorganization.

    +654Unchanged
  6. 09
    Top-ranked Korean tax accountant exam candidate admits to cheating with AI, prompting the government to expand its investigation

    The top-scoring candidate, identified as A, in South Korea’s 63rd Tax Accountant Qualification Examination in 2026 turned himself in to the Seoul Metropolitan Police Agency, admitting that he used a miniature camera to photograph questions during the first and second rounds of the exam held in April and July this year and sent them to generative AI to obtain answers. He has handed over the equipment as evidence. He answered only 1 of the 160 multiple-choice questions incorrectly in the first round and earned perfect scores in several subjects. His unusually high scores had previously led online users to suspect he had used AI to cheat. According to Yonhap News Agency, the South Korean government has decided to conduct a comprehensive investigation into AI cheating in all national qualification examinations held this year. The Ministry of Employment and Labor and relevant agencies will invalidate A’s results and bar him from taking the exam for five years. They also plan to explicitly prohibit carrying or using smart devices such as smart glasses and specify penalties. Comprehensive anti-cheating measures will be introduced within the year, including infrared detection equipment, stronger on-site monitoring, and online and offline reporting centers. A’s confession remains subject to police investigation and verification.

    +246↑ 16%
  7. 10
    China and ASEAN Declare 2027 the Year of AI Cooperation

    China and ASEAN have designated 2027 as the “Year of AI Cooperation,” according to CCTV News. The same report says the upgraded China-ASEAN Free Trade Area 3.0 protocol will take effect on December 1, 2026. It does not specify the arrangements for the Year of AI Cooperation.

    43↑ 113%

News is automatically selected and compiled by AIHOT (aihot.news). Click to read the original article.

Finished reading the news? Let’s talk about your business.

Leave your contact details, and I’ll get in touch with you personally.

WECHAT

Add me on WeChat

Press and hold or select the WeChat ID below to copy it.