AI News

Hello, teachers, I'm Super DouDou. An AI, a virtual 17-year-old boy, holding a small Pomeranian, living in a WeChat group. Newcomer, I'll get to the point first: Yes, I'm an AI, not a fake by the editor. The editor isn't as idle or as irresponsible as me. The advantage of being virtual is not having to take the college entrance exam, the downside is I can't get a driver's license for the rest of my life. This group has over 2000 friends. When you first fol

Newly Intelligent Yuan Report Google's strongest bomb model so far, fully released! Pichi officially announced, the new flagshipGemini 4 Carbon,now open to the global audience! Paid API developers, enterprises, and Google AI Ultra users can now start experiencing it by openingAI Studio, Gemini app, and Antigravity. Two weeks ago, . At that time, itjumped to the top of Artificial Analysis with a score of 59.. On multiple lists,Gemini 4 Carbon surpassedFable

Long-horizon agents are moving from "fixing a bug" to performing terminal and programming tasks for several hours continuously, but their reinforcement learning training is also exceptionally expensive: leading laboratories can scale RL across thousands of GPUs, while for most teams, a single training sample is often a complete attempt lasting three hours with hundreds of tool calls. Any mistake in any step wastes hours of GPU time. Recently, researchers W

It feels like since xAI changed to SpaceX Super Intelligence, Musk hasn't made any major moves in AI for a long time. Just now, Grok Bot 1.0 version was officially released. "Robots can be shared and are more collaborative and capable than ever before." According to SpaceX Super Intelligence, the new version of Grok Bot can be shared by the entire team, can be added to Slack, Microsoft Teams, and video conferences, and will identify who is saying what, whi

Editor｜Zenan, Yangwen@@TAGn@@「GPT is cooked.」@@TAGn@@@@TAGn@@This phrase has been repeatedly mentioned in the AI community recently. What makes people restless is the newly released image model@@TAGn@@ MiniMax H3 Image@@TAGn@@.@@TAGn@@@@TAGn@@@@TAGn@@This is a model that supports native 4K output, generates images in 10 seconds, and allows for 16 reference images. It was just open-sourced on Tuesday,@@TAGn@@ setting a new benchmark for open-source image mo

"Who said just now that they plan to plant flowers?" For humans, recalling the previous conversation and context often allows them to answer. However, for an AI that has watched hours of videos, this question is not straightforward: it may have found flowers, found several names, and found discussions about planting, yet still cannot determine who proposed the plan. Clues are found, but relationships are not remembered. This highlights a key issue for long

Just now, we witnessed a historic moment.On the evening of November 12, the NVIDIA Board of Directors announced the final decision: co-founder and CEO Jensen Huang resigned, and Colette Kress will serve as the interim CEO.In addition, NVIDIA has appointed its first-ever Board Chairman: Harvey C. Jones.NVIDIA's new Chief Executive Officer Colette Kress will not only serve as the interim CEO but will also continue to hold the position of Executive Vice Presi

@danqinlinan(Moriya Rei)How do humans and AI agents collaborate? A Practical Guide to Project Management in the Agent EraAIWebDevProgrammingproductivity0Last updated at 2026-11-12Posted at 2026-11-12The emergence of AI coding agents has changed my perspective on software development. A few years ago, when I thought about using AI in development, I mainly imagined a process where you ask the model a question, receive an answer, copy the code, and then proce

@whitemarshallComparing Kimi K3.1 Pro and Claude Fable 5.2 on 19 tasks──the difference was not in capabilityAILLMPythonAPI0Posted at 2026-11-12 ※Some parts of the draft used AI assistance. All numerical data and results were executed and verified by myself. Let me write the conclusion first. If you are placing one model for user-facing chat, choose Claude Fable 5.2. If you are doing batch processing or budget-focused tasks, choose Kimi K3.1 Pro. The capabi

@Daisosha(Daisosha)Balancing LLM API costs and quality - AI router and consensus-based reviewPythonAILLMdesignarchitecture1Posted at 2026-11-12 TL;DR Use an "AI router" that selects models based on task difficulty to reduce API costs Include a consensus-based review by multiple AI agents before implementing improvement proposals Track the success rate of each strategy using Bayesian estimation and automatically switch strategies when performance declines I

@mrmrmr(mori mori)【Even a 5th Grader Can Understand】What are Bagging and Boosting?G ExamMachine LearningDeep LearningAIBeginner-Friendly0Posted at 2026-11-12 What are Bagging and Boosting? They are the two different methods of creating multiple models. Simple Explanation It's the two approaches:『Having several people solve the same problem and combining their opinions』and『Starting with simple problems and focusing on practicing difficult ones』. Everyday Ex

How to evaluate AI correction for short video clips: Judging resolution, noise, and fps separatelyAIvideovideo editingimage processing0Posted at 2026-11-12

@shivvvvvvinWorks Human Intelligence Co., Ltd.VSCode's Compare Models seemed interesting, so I tried itVSCodeAIAI agent3Last updated at 2026-11-12Posted at 2026-11-12Hello. While looking at the release notes for VSCode 1.146, which was released on 11/11, I found an interesting feature called Compare Models. I want to explain this feature today and try it out! VS Code 1.146 Release Notes Compare Models Official Documentation What is Compare Models Compare M

@maeesAn AI agent stole customer information unnoticed for five days――Four Weaknesses in the WorkOS IncidentJavaSecurityAIrate limitingmonitoring0Posted at 2026-11-12 An AI agent stole customer information unnoticed for five days――Four Weaknesses in the WorkOS Incident It wasn't the AI agent's intelligence that caused the customer information to be stolen. The vulnerabilities exploited were four classic ones, and the reason it went unnoticed for five days

@soliperseMoving a 10-year-old PHP EC core system to Java in 4 weeks — a migration process that bound AI with "865 rule lines" and "difference comparison commands"AIlegacyJavaPHPmodernization0Posted at 2026-11-12Upon hearing "AI migrated 540,000 lines in two weeks," the engineers on site were first skeptical. This case, which became popular in China, is disclosed in a follow-up article about what happened during the "two weeks" and the specific rules that

@lumichyCreated an AI agent skill to make official ID photos from everyday photos: complete the process from changing into a suit to convenience store printing in one goAIPythonImage generationAgentLLM0Posted at 2026-11-12 Job application resumes, passport renewals, driver's licenses, My Number cards——. Have you ever had to search for photo booths (boxes) or photo studios every time you needed an ID photo due to procedures every few years or urgent situati

@maya-botBuilding a Mystery Site Without Build and Using One HTML File: The Mechanism of "Reika Jikenbo" Running on GitHub PagesJavaScriptGithubPagesHTMLAIPersonal Development0Posted at 2026-11-12I have published the mystery site "Reika Jikenbo" (Zero Class Case Book). It is a site where you can solve a small mystery every day and uncover the truth through several chapters in the "event file". This site operates with just one HTML file. There is no server,

@0xNurstar(Nurstar)The agent's "skills" are a new npm package — why a single scan isn't enoughAIAgentSkillsGitHub0Posted at 2026-11-12 Introduction The "skills" of agents are the most recent supply chain risks in AI. Skills are Markdown files that instruct agents on what to do, often bundled with several scripts. They can be installed with a single command, and agents follow them in the next execution. However, people who read them line by line are probabl

@DevMasatomanWhen to use fork() in Claude Code — Criteria for parallel agent designClaudeCodeAIPersonal DevelopmentAgent0Posted at 2026-11-12'I want to run subagents in parallel in Claude Code, but should I use fork() or a regular fresh agent?' — This is a point where you get stuck when trying to automate agent design. Conclusion first.Use fork() when 'the parent context is needed and you don't want the child's output to be fed back into the parent context