Welcome to the 【AI Daily】 section! Here is your guide to exploring the world of artificial intelligence every day. Each day, we present the hottest topics in the AI field, focusing on developers to help you gain insights into technological trends and understand innovative AI product applications.
Fresh AI Products Click to Learn More: https://top.aibase.com/
1. DeepSeek's Liang Wenfeng May Become Asia's Top Tech Tycoon with a Valuation of Up to $150 Billion
Liang Wenfeng, the founder of DeepSeek, may become one of Asia's top tech tycoons due to his success in the artificial intelligence field. According to Bloomberg's research, DeepSeek's valuation is expected to be between $1 billion and $150 billion. With an 84% stake, Liang could join the ranks of Asia's wealthiest tech magnates. Although DeepSeek has not accepted external financing, it has achieved financial independence through profits from a quantitative hedge fund.
【AiBase Highlights:】
💹 DeepSeek's valuation is estimated between $1 billion and $150 billion, with Liang Wenfeng potentially holding $126 billion in assets.
🤖 DeepSeek has not accepted external financing and has achieved financial independence through the profits of the quantitative hedge fund "Phantom Quant."
🏆 Liang Wenfeng is poised to become one of Asia's top tech tycoons, possibly surpassing Nvidia CEO Jensen Huang.
2. ByteDance and HKU Release New Video Model Goku: Can Generate Virtual Digital Human Videos Directly
Recently, the Goku video generation model developed by the University of Hong Kong in collaboration with ByteDance was officially released. This model is based on advanced generative algorithms and can generate high-quality video content from text prompts, greatly enriching the forms of digital art. The research team showcased a series of impressive video examples, demonstrating Goku's unlimited potential in creative expression.
【AiBase Highlights:】
🌟 The Goku model was developed in collaboration with the University of Hong Kong and ByteDance, capable of generating high-quality video content from text prompts.
🎨 The model showcases various scenes, including a fashionable woman strolling through the streets of Tokyo and a giant mammoth walking in the snow, with vivid and realistic effects.
💡 The release of the Goku model provides new tools for visual art creation, helping creators explore more possibilities.
Details link: https://saiyan-world.github.io/goku/
3. Musk Seeks to Acquire OpenAI for $97.4 Billion; Altman Responds Humorously: We Can Also Acquire Twitter
Recently, Tesla CEO Elon Musk proposed to acquire control of OpenAI for $97.4 billion but was humorously rejected by CEO Sam Altman. Musk hopes that the acquisition will help OpenAI return to its original startup vision, emphasizing that artificial intelligence should benefit humanity. However, under Altman's leadership, OpenAI has gradually transformed into a for-profit company, believing that commercialization is key to driving technological development, which is significantly different from Musk's original intention.
【AiBase Highlights:】
🌟 Musk proposed to acquire OpenAI for $97.4 billion but was rejected by CEO Altman.
🤝 Musk hopes to help OpenAI return to its original purpose, emphasizing that technology should be beneficial to humanity.
📈 OpenAI is transitioning to a for-profit company, and Altman believes commercialization is key to driving technological development.
4. Reports Indicate OpenAI Will Complete Its First Self-Developed Chip, TSMC to Handle Foundry
OpenAI is accelerating the design of its first artificial intelligence chip, aiming to reduce reliance on Nvidia. Large-scale production is expected by 2026, with TSMC responsible for wafer testing. This progress marks a significant breakthrough for OpenAI in the chip development field and lays the foundation for future self-sufficient chip production. Despite the technical challenges of wafer testing, the OpenAI team is well-prepared to ensure the smooth testing of the chip.
【AiBase Highlights:】
🌟 OpenAI is actively advancing the design of its self-developed AI chip, aiming to reduce reliance on Nvidia chips.
🏭 TSMC will handle the wafer testing for the chip, with large-scale production expected in 2026.
🔧 OpenAI is prepared to tackle technical challenges to ensure smooth chip testing, laying the groundwork for future development.
5. ZyphraAI Launches Open Source Zonos-TTS: Supports Real-Time Voice Cloning and Multilingual Emotion Control
ZyphraAI recently launched its new multilingual text-to-speech model Zonos-TTS, marking another advancement in speech synthesis technology. The model is not only open-source but also commercially usable, supporting real-time voice cloning and trained on extensive English voice data, showcasing excellent performance. Users can choose local deployment or API services, and it offers free audio generation services, significantly lowering the entry barrier.
【AiBase Highlights:】
🎤 The Zonos-TTS model is based on the Apache 2.0 license, fully open-source and commercially usable.
💻 It offers local deployment and API services, allowing regular users to generate 100 minutes of audio for free each month.
⚡ It runs at about twice real-time speed on an RTX 4090 graphics card, demonstrating high efficiency.
Details link: https://huggingface.co/Zyphra/Zonos-v0.1-hybrid
6. Alibaba Tongyi Lab Releases Open Source Music Generation Technology InspireMusic
Tongyi Lab recently launched InspireMusic, an open-source music generation toolkit designed to provide researchers, developers, and music enthusiasts with a comprehensive creative platform. This toolkit not only offers a variety of music generation models and training tools but also lowers the barriers to music creation, allowing users to generate diverse musical works through simple text descriptions or audio prompts.
【AiBase Highlights:】
🎵 InspireMusic is an open-source music generation tool suitable for researchers and music enthusiasts.
🎹 The toolkit supports various musical styles and emotional expressions, allowing users to generate music through text descriptions.
🎧 InspireMusic provides flexible reasoning modes to meet different user needs for generation speed and audio quality.
Details link: https://github.com/FunAudioLLM/InspireMusic
7. Zhihu AI Search Product Zhihu Direct Answer Integrates "Full-Strength" DeepSeek-R1
Recently, Zhihu announced a major upgrade to its AI search product Zhihu Direct Answer, integrating the DeepSeek-R1 model to enhance reasoning capabilities and user experience. DeepSeek-R1, known for its excellent logical deduction and structured output capabilities, combined with Zhihu's vast high-quality corpus, will significantly improve the accuracy and quality of AI search results. 
【AiBase Highlights:】
🚀 Integrating the DeepSeek-R1 model enhances Zhihu Direct Answer's reasoning abilities and search experience.
📚 Combined with the community's high-quality corpus, it provides more accurate and structured search results.
💡 A new knowledge base feature is added to help users manage personal knowledge and improve information processing efficiency.
8. Baidu Wenxiaoyan App Update: Introduces DeepSeek-R1 Model, Significantly Enhancing Photo Problem-Solving Ability
Baidu Wenxiaoyan App has recently been updated to version 4.9.0, significantly enhancing the photo problem-solving feature, allowing users to clearly see the thinking steps involved in the problem-solving process. Additionally, the new version introduces a personalized companion learning mode and creative tools, enhancing the learning experience. This update marks a deeper application of artificial intelligence in the education field, with expectations for further optimization of features to better serve learners.
【AiBase Highlights:】
📸 The photo problem-solving feature has been significantly enhanced, allowing users to view problem-solving thought processes more clearly.
🎓 A companion learning mode has been added to provide users with a personalized learning experience, enhancing learning enjoyment.
📝 The introduction of handwritten reports and essay functions encourages students' creativity and writing skills.
9. OPPO AI Integrates DeepSeek-R1; Find N5 to be Released on February 20
OPPO announced today that it will release its latest foldable smartphone, Find N5, on February 20. This phone boasts the title of the world's thinnest foldable screen, with a single-side thickness of only about 4mm and a thickness of less than 9.2mm when folded. Additionally, the Find N5 is equipped with the Qualcomm Snapdragon 8 Gen 2 chip and features a 3D-printed titanium alloy hinge, supporting IPX6, IPX8, and IPX9 full-level waterproofing. Furthermore, Find N5 integrates the DeepSeek-R1 open-source reasoning model, providing a more intelligent human-computer interaction experience.
【AiBase Highlights:】
📏 Find N5 is the world's thinnest foldable smartphone, with a single-side thickness of about 4mm and a thickness of less than 9.2mm when folded.
💪 It features the Qualcomm Snapdragon 8 Gen 2 chip and a 3D-printed titanium alloy hinge, supporting full-level waterproofing (IPX6, IPX8, IPX9).
🤖 Integrating the DeepSeek-R1 model allows users to enjoy an intelligent interaction experience, supporting voice assistant and online search functions.
