Welcome to the 【AI Daily】 column! This is your daily guide to exploring the world of artificial intelligence. Every day, we present you with hot topics in the AI field, focusing on developers, helping you gain insights into technological trends and innovative AI product applications.
Fresh AI Products Click to Learn More: https://top.aibase.com/
1. ByteDance Doubles Down on AGI: Doubao Large Model Forms "Seed Edge" Team to Explore the Future of AI
ByteDance is increasing its investment in fundamental research in artificial intelligence by forming the "Seed Edge" team, which focuses on long-term AGI research. The team has identified five major research directions, demonstrating ByteDance's strategic commitment to foundational theory and technology development in AI. The goal is not just to apply AI but to achieve fundamental breakthroughs in AI technology, laying the groundwork for future AGI development.
【AiBase Highlights:】
🚀 ByteDance has established the "Seed Edge" team, focusing on long-term AGI research, showcasing its strategic commitment in the AI field.
🔍 The team has identified five major research directions, covering reasoning and perception capabilities, model design, learning paradigms, etc., delving into foundational theory and technology development.
🎓 ByteDance is recruiting top research talent through the "Top Seed Talent Program" and collaborating with universities to enhance its capabilities and determination in AI foundational research.
2. Based on CogAgent! Zhipu GLM-PC Computer Intelligent Agent Large Model Open Experience
GLM-PC is a computer intelligent agent based on CogAgent, equipped with powerful logical reasoning and deep perception capabilities. It supports Windows systems and enhances code generation and complex task processing through the latest "Deep Thinking" mode. The design of GLM-PC is inspired by the division of labor between the left and right hemispheres of the human brain, enabling efficient task execution and excellent user interaction experience.
【AiBase Highlights:】
🧠 GLM-PC enhances logical reasoning and code generation functions through the "Deep Thinking" mode, improving the ability to handle complex tasks.
🛠️ Its architectural design draws on the division of labor between the left and right brain, achieving a deep integration of logic and creativity, optimizing task execution and perception.
🤝 Zhipu actively promotes collaboration with well-known PC manufacturers to drive innovation in AI personal computers and enhance user experience.
Details link: https://cogagent.aminer.cn
3. Step-Video V2 Model Released by Jumpshot Star
On January 22, 2025, Shanghai Jumpshot Star Intelligent Technology Co., Ltd. released the Step-Video V2 version, marking a significant advancement in video generation technology. This version optimizes the VAE model and DiT architecture, significantly improving generation efficiency and video quality. Additionally, the newly added basic text generation feature enriches video content and broadens application scenarios.
【AiBase Highlights:】
🎥 Step-Video V2 uses an efficient VAE model to reduce computational complexity and enhance generation efficiency.
🤖 By deeply optimizing the DiT architecture and introducing reinforcement learning, it improves the fluency and detail expressiveness of video generation.
📝 The new basic text generation feature allows for natural integration into video content, expanding application scenarios.
Details link: https://yuewen.cn/videos
4. Google Gemini AI Assistant Major Upgrade: Can Analyze Images and Control Multiple Applications!
Google has recently conducted a comprehensive upgrade of its Gemini AI assistant, particularly enhancing multimedia processing capabilities and application interactivity for Android devices. The newly introduced "Gemini Live" feature allows users to directly insert images, files, and videos during conversations with the chatbot, enhancing interactivity. The upcoming Project Astra will also introduce screen sharing and real-time video streaming capabilities, further enhancing Gemini's intelligence level.
【AiBase Highlights:】
🌟 The Gemini AI assistant adds the "Gemini Live" feature, allowing users to insert images, files, and videos during conversations.
📱 Gemini's expanded capabilities enable it to collaborate with more applications, allowing users to consolidate multiple tasks.
🔍 The upcoming Project Astra will introduce screen sharing and real-time video streaming capabilities for Gemini.
5. Musk Slams Trump's $500 Billion AI Investment Plan as a "Financial Farce"
Recently, Trump announced an artificial intelligence investment plan called "Gateway to the Stars," with a budget of up to $500 billion, jointly funded by OpenAI, Oracle, and SoftBank. Despite the widespread attention this plan has garnered, Elon Musk, CEO of Tesla and SpaceX, expressed skepticism, calling it a financial "farce" and questioning the authenticity of the funding.
【AiBase Highlights:】
🌟 Trump announces the $500 billion "Gateway to the Stars," funded by OpenAI, Oracle, and SoftBank.
💬 Musk questions the plan, calling it a financial "farce" and doubting the real existence of the funds.
🤝 Altman responds to Musk's remarks, expressing respect for his achievements but also delivering a rebuttal.
6. Claude Founder Predicts: AI Will Fully Surpass Humans in 2-3 Years
In a recent interview, Anthropic CEO Dario Amodei expressed optimism about the future development of artificial intelligence, predicting that AI systems will surpass humans in multiple fields within the next 2-3 years. He emphasized that AI development should complement humans to enhance productivity rather than simply replace them. He also encouraged young people to learn to use AI tools and cultivate critical thinking to cope with the era of information explosion.
【AiBase Highlights:】
🚀 Amodei predicts that AI systems will surpass humans in multiple fields within the next 2-3 years.
🧠 He emphasizes that AI should complement humans to enhance productivity rather than replace human jobs.
📈 Claude will introduce new features to enhance its applications in enterprise-level services.
7. 18-Year-Old Earns $56 Million with ChatGPT
Amid the wave of artificial intelligence, young people are achieving financial freedom through innovation. Zach Yadegari and Blake Anderson developed the Cal AI app using ChatGPT, successfully generating $56 million in revenue. The app quickly attracted a large user base with its unique feature of identifying food calories through photos and effective marketing strategies.
【AiBase Highlights:】
📱 Cal AI is a calorie tracking app that identifies food calories through photos, allowing users to upload pictures for detailed analysis.
🚀 The app achieved a rapid increase in brand awareness due to its 90% accuracy rate and effective marketing strategies, including collaborations with KOLs and promotions on social platforms.
💵 A user-centered pricing strategy, with a monthly fee of only $10 and an annual fee of $30, allowed Cal AI to quickly accumulate a large user base and achieve significant commercial returns.
8. Samsung Galaxy S25 First to Support Content Credential Standards, Aiding in Identifying AI-Generated Images
At Samsung's recent Unpacked event, the Galaxy S25 series was announced as the first smartphone to support content credential standards. This standard, proposed by Adobe, aims to provide "nutrition label" information for digital content, helping users identify AI-generated content.
【AiBase Highlights:】
📱 The Galaxy S25 series is the first smartphone to support content credential standards, aimed at identifying AI-generated content.
🔍 Content credentials provide "nutrition labels" for digital content, revealing the generation and editing processes of the content.
🤝 Members of the C2PA alliance include renowned companies such as Samsung, Adobe, and Google, working together to combat misinformation.