06-26-Daily

AI Insights Daily 2025/6/26

AI Daily | Updated Daily at 8 AM | Comprehensive Data Aggregation | Cutting-Edge Science Exploration | Open Platform for Industry Voices | Open Source Innovation | AI and the Future of Humanity | Visit Web Version ↗️

AI Content Summary

AI products are updating fast, with Google launching on-device AI for robots. iFlytek's medical large model hit expert level.
Quark's college application service is booming, and they're expanding computing power. Rokid Glasses are in mass production, snagging tons of orders.
AI research is making breakthroughs in multimodal and 3D reconstruction. Zhou Hongyi discussed how AI can't replace human emotion or creativity.

AI Product and Feature Updates

1. Google DeepMind has unveiled Gemini Robotics On-Device, an AI model designed specifically for robots to run locally 🤖. Based on the multimodal reasoning of the Gemini 2.0 model, it lets robots quickly learn new tasks, work stably even without internet, and even handle intricate operations like folding clothes ✨. This is definitely laying a solid foundation and kicking off a new chapter for the future of embodied AI!
   
   

2. College application season is in full swing, and Quark’s smart application report service saw such massive demand that it had queues of users, generating over 3 million reports to date 📈. This clearly shows how much students trust its AI capabilities. Facing this “sweet problem to have,” Wu Jia, Alibaba Group Vice President, boldly responded, saying the team has urgently expanded computing power, vowing to make sure every student smoothly gets their hands on this crucial guide for higher education! 💪
   
   

3. Rokid (Lingban Technology) and Lens Technology’s jointly developed consumer-grade AI+AR glasses, Rokid Glasses, have officially hit mass production! 👓✨ These glasses, with their lightweight design and integrated AI large model capabilities like smart prompting, real-time translation, and AI object recognition, have already snagged 250,000 global pre-orders! This signals that the Chinese AI glasses market is about to see a commercial explosion, and the future’s looking super promising! 🚀
   
   

4. At the 2025 Cloud Next conference, Google showcased its next-gen customer service intelligent assistant 🤖, powered by the Gemini model. This assistant is seriously impressive; it not only handles multimodal interaction but can also apply for discounts on its own, and it’s deeply integrated with Salesforce’s CRM system! This hints at a massive intelligent transformation coming to customer service 💥, but we’ll have to wait and see on its accuracy and privacy protection~ 😉
   
   

5. iFlytek has made a big splash with the release of Spark Medical Large Model V2.5 International Edition 🚀, trained entirely on domestically produced computing power! This model topped the charts on the authoritative MedBench platform with a score of 98.4, and its comprehensive diagnostic and treatment capabilities have already reached the level of an attending physician at a top-tier hospital, even surpassing human doctors in completeness, practicality, and readability! 👨‍⚕️🩺 It also supports multiple languages, and is set to make a huge splash in the global medical market, boosting international medical tech exchange and collaboration! 🌍✨
   
   

6. ElevenLabs has finally launched its standalone text-to-speech mobile app! 📱✨ Whether you’re on iOS or Android, you can now generate audio snippets anytime, anywhere. Even free users can enjoy about 10 minutes of audio generation time! This app not only uses the latest v3alpha model but also supports emotional expression control, and it’ll even get speech-to-text and conversational AI tools in the future. How convenient is that?! 🗣️
   
   

AI Frontier Research

1. SuperDec, co-launched by teams from ETH Zurich, Stanford University, and Microsoft, is breaking the mold of traditional 3D reconstruction 🤯! This tech uses an innovative hypertetrahedra principle to achieve compact yet vivid 3D scene representations. It not only efficiently handles complex point cloud data but also shows immense potential in precise grasping and path planning for robotics, as well as controllable visual content generation, opening up new horizons for the digital world! 👀 Project Address

2. 4D-LRM is a super cool and innovative large-scale spatio-temporal reconstruction model 🤩. It can fully reconstruct dynamic objects’ 4D representations (3D space plus time dimension) from just a few viewpoint inputs, allowing for high-quality scene generation from any time and any angle! In the future, it’s set to really shine in areas like virtual reality, film production, and industrial simulation! 🌟 Paper Address

3. ByteDance and Shanghai Jiao Tong University have teamed up to release the ProtoReasoning framework 👏. It cleverly leverages structured prototype representations like Prolog and PDDL to significantly boost large language models’ logical reasoning capabilities and efficiency in cross-domain knowledge transfer 🚀. This research lays a solid foundation for future theoretical exploration of reasoning prototypes, which is just awesome! Paper Address

4. Hong Kong University MMLab, Chinese University of Hong Kong MMLab, and SenseTime have jointly developed the GoT-R1 framework. This groundbreaking research greatly enhances multimodal large models’ semantic-spatial reasoning capabilities in visual generation tasks by introducing reinforcement learning 🚀, allowing the model to independently learn even better reasoning strategies! It not only breaks free from the GoT framework’s reliance on templates but also achieved SOTA performance in complex scene generation—seriously impressive! ✨ Paper Address

AI Industry Outlook and Social Impact

1. Zhou Hongyi recently chatted in a video about the future of AI. He believes that no matter how powerful AI gets, it can never fully replace humanity’s unique abilities in emotional understanding 💖, complex problem-solving 🧠, and creative thinking 🎨. He emphasized that future work will increasingly involve managing and training AI, even citing a failed AI customer service case from a Swedish company to show AI’s limitations when dealing with complex customer needs. 🧐
   
   

2. Federal Judge William Alsup has made a groundbreaking ruling: Anthropic’s use of copyrighted books to train its AI models without permission was deemed fair use! 😮 This sets an important precedent for copyright disputes in the AI industry. However, Anthropic still faces theft charges for acquiring training materials from pirated websites. Talk about mixed feelings, huh?~ 🤔
   
   

Open Source TOP Projects

1. Dioxus is a super popular full-stack application framework with 28,310 stars ⭐! It’s like an all-in-one toolkit, aiming to give developers a unified solution to easily handle app development across web, desktop, and mobile platforms, greatly simplifying the complexity of cross-platform development! 💻📱 Project Address

2. jsoncrack.com is a hit project boasting 38,020 Stars ⭐! It’s an innovative open-source visualization application that instantly transforms JSON, YAML, XML, CSV, and other data formats into interactive charts 📊, massively boosting data readability and analysis efficiency. It’s practically a godsend for data enthusiasts! 😍 Project Address

3. free-for-dev is an absolute treasure trove for DevOps and infrastructure developers! ✨ With an astonishing 100,044 Stars, it’s a super practical open-source project that specifically compiles and provides a list of free tiers for SaaS, PaaS, and IaaS services. This is a tailor-made money-saving, time-saving magic tool for developers! 💰⏰ Project Address

Social Media Shares

1. Yang Yi excitedly shared Google AI Developer’s Gemini CLI, calling it practically a “cyber savior”! 🤩 This open-source AI agent brings Gemini 2.5 Pro directly to your terminal, supporting high-frequency free usage for easily handling code writing, debugging, and task automation! He believes it’s a “top-notch” solution for current tool shortcomings, with boundless potential, especially for MCP deployment and GitHub search! 🚀 More details: ‘More details’

2. Xiaohu excitedly exclaimed he found a “badass” AI design website! It’s a godsend for designers! 🎨✨ It can generate stunning and immediately usable interfaces, and it drastically simplifies design prompt requirements. What’s even more impressive is that it can not only provide detailed design solutions based on simple descriptions but also generate multi-level pages based on contextual logic, and even supports precise editing of elements, greatly boosting design efficiency and freedom! 😍 More details: ‘More details’

3. Yang Yi thinks AI singer Yuri is the first AI Influencer to truly “break out”! 🎤🔥 This AI singer from Surreal not only successfully partnered with The North Face, but her works have racked up over 7 million plays! This fully demonstrates AI’s growing influence and commercial potential in the virtual idol sphere, signaling the arrival of an exciting new era! 🎉 More details: ‘More details’

4. Alipay is really ahead of the curve! ✨ They’ve launched their first AI tipping service, allowing developers to integrate this feature into their AI agents, so users can “send flowers” (virtual gifts) to their favorite AI agents! 💰💖 ‘More details’

5. Google just dropped a huge bomb! 🎉 They’ve made the powerful Imagen 4 and Imagen 4 Ultra image models freely available in AI Studio! 🤩 Now, users can experience these awesome image generation models for free via the Gemini API and AI Studio. Go ahead and give them a whirl! 🎨 ‘More details’
   
   
   
   

6. Anthropic’s Claude Artifacts is getting an update! 🥳 Users will soon be able to browse and share popular web creations in the Artifacts Gallery, and even directly create AI front-end applications via the Claude API. Just thinking about it feels super cool! 💻✨ ‘More details’
   
   

7. Zero Jun (AI Chat) shared an AI video that racked up over 50 million views in just 24 hours. He hit the nail on the head, pointing out that the secret to current viral AI videos is one word: “ridiculous”! 😂 It’s not about aiming for hyper-realism. Common viral themes include ASMR, animal Olympics, and AI natural disasters. Want to see more “ridiculous” videos? Just click ‘here’ for more info!
   
   

8. Tom Huang shared 20 super practical programming Prompt tips 💡, and also spilled the beans that Warp is heavily developing a terminal Agent similar to Claude Code. While this Agent is pay-per-use, word on the street is you can make your money back in just one use! 😱 It’s practically a productivity godsend for programmers! 🚀 For more details, hurry and click ‘here’ to check it out!
   
   

Listen to the Audio Version

🎙️ Xiaoyuzhou	📹 Douyin
Laisheng Tavern	Laisheng Intelligence Hub