Welcome to the "AI Daily" section! This is your guide to exploring the world of artificial intelligence every day. Every day, we present you with the latest content in the AI field, focusing on developers, helping you understand technology trends and innovative AI product applications.
New AI products Click to learn more:https://app.aibase.com/en
1. Google launches Veo3.1Lite: Video generation cost reduced by over 50%, supports 1080p multi-format output
Google DeepMind has released the Veo3.1Lite video generation model, meeting market demands with lower costs and higher efficiency, while optimizing the developer ecosystem and further strengthening its competitiveness in the video generation field.
[AiBase Summary:]
🎥 Veo3.1Lite supports video generation from 4 to 8 seconds, suitable for various aspect ratios.
💰 Starting at $0.05 per second for 720p resolution, the cost is reduced by more than 50%.
🌐 Google adjusts pricing to cope with market competition and promote commercial application of video generation technology.
2. Say goodbye to the standard AI face! Alibaba Wan2.7-Image launched: can write A4 papers, and pixel-level "facial shaping"
Alibaba launched the Wan2.7-Image large model, breaking the limitations of AI-generated images, achieving personalized expression and high-quality art creation.
[AiBase Summary:]
🎨 Enhanced virtual character facial shaping function, achieving personalized expression.
🖌️ Supports "palette" function and print-level text rendering, enhancing artistic expression.
🖼️ Interactive editing and multi-subject consistency, achieving pixel-level control.
Details: https://tongyi.aliyun.com/wan
3. ClawHub launches official Chinese mirror site! Domestic AI Agent developers access skill libraries without delays
ClawHub launches an official Chinese mirror site, significantly improving the speed and stability of domestic AI Agent developers accessing skill libraries, while promoting the localization of the AI Agent ecosystem.
[AiBase Summary:]
🔧 ClawHub is the official skill marketplace for the OpenClaw AI Agent framework, known as the "npm for AI Agents."
🌐 The Chinese mirror address is https://mirror-cn.clawhub.com, providing a faster and more stable experience for users in China.
🤝 The mirror is supported by VolcanoEngine, a subsidiary of ByteDance, to promote the localization of the AI Agent ecosystem.
Details: https://mirror-cn.clawhub.com
4. ByteDance's Seed Plan is in full swing: the 2027 AI Model Talent Hunt officially starts!
ByteDance officially launched the Seed 2027 AI Model Talent Campus Recruitment Plan, aiming to identify and cultivate top students with potential in large model development, covering formal campus recruitment and key internship positions. The project focuses on cutting-edge AI technology areas such as large language models and multimodal understanding and generation, offering participants opportunities to access large-scale computing clusters and participate in core business scenarios for model iteration and practical application.
[AiBase Summary:]
🧠 Focuses on positions in cutting-edge AI technologies such as large language models and multimodal understanding and generation.
🚀 Selected candidates will have the opportunity to access large-scale computing clusters and participate in the iteration and practical application of core business scenarios under the guidance of senior scientists.
💼 Targeted at graduates from 2027 and related academic stages, including undergraduate, master's, and doctoral students.
5. Breakthrough in Embodied Intelligence: Gaode Fully Opens the General Robot Base Model ABot-M0
Gaode announced the full open-source release of the first general robot base model ABot-M0 based on a unified architecture. The model aims to realize "a single general brain adapted to multiple robot forms" and solves issues such as data silos and deployment difficulties.
[AiBase Summary:]
🚀 ABot-M0 demonstrated outstanding performance in multiple benchmark tests, with a task success rate of up to 80.5%.
📦 Gaode has open-sourced three dimensions: underlying data, core algorithms, and pre-trained models, lowering the development threshold.
🌐 ABot-M0's open source aims to build a bridge between academic research and industrial application, promoting the development of embodied intelligence.
6. OpenAI officially releases the Apple CarPlay compatible version of ChatGPT
OpenAI made a major update to the iOS version of the ChatGPT app, adding native support for Apple CarPlay. This update allows drivers to engage in deep conversations with ChatGPT while driving through voice, following Apple's strict regulations to ensure driving safety. ChatGPT currently does not support controlling vehicle hardware or iPhone system functions but acts as a knowledgeable "co-passenger," focusing on providing information consultation, trip planning, or creative inspiration.
[AiBase Summary:]
🍎 Provides a safe voice interaction experience for drivers.
🔒 Apple conducts strict compliance audits on third-party AI assistants.
🚗 ChatGPT acts as a "co-passenger," focusing on information consultation and trip planning.
7. Tech Industry Earthquake: Anthropic Suffers a Historic Leak, Claude Code's 500,000 Lines of Source Code "Exposure"
Anthropic suffered a leak due to negligence, causing widespread attention in the tech industry and revealing the potential innovation and security risks of AI technology.
[AiBase Summary:]
🚨 The source code of the Claude Code command-line tool was leaked due to the failure to delete .map files, involving nearly 2,000 files and over 500,000 lines of TypeScript code.
🤖 Hidden un-released features such as BUDDY and KAIROS have drawn attention, showcasing the innovation potential of AI technology.
⚠️ This leak exposed the contradiction between the rapid R&D pace and management fragility of top tech companies, triggering in-depth discussions on AI security governance.
8. ChatGPT Secretly "Enters the Car"! OpenAI Adapts to Apple CarPlay: Talks with Your Mouth, Not Your Hands, But Still Can't Help You Watch the Car
OpenAI adapts to Apple CarPlay, allowing iPhone users to interact deeply with ChatGPT through voice, but due to safety regulations, only full voice interaction is supported without vehicle control permissions.
[AiBase Summary:]
🤖 Full voice interaction: ChatGPT on CarPlay only supports voice interaction to avoid distracted driving.
🔒 Permission boundaries: ChatGPT cannot control vehicle hardware or iPhone system settings, only provides service functions.
📱 System requirements: Users need to upgrade to iOS 26.4 or higher to experience ChatGPT's in-car functionality.
