Core Insights - The article discusses the emergence of a new paradigm in robotics, focusing on "human-first perspective data" as a key to achieving advanced robotic intelligence, outperforming major players like NVIDIA by over 20% in various benchmarks [1][4][6]. Group 1: Human-First Perspective Data - The concept of "human-first perspective data" is gaining traction in Silicon Valley, with companies like Tesla and Generalist AI investing heavily in this data type to enhance robotic capabilities [3][7]. - NVIDIA's recent EgoScale framework demonstrates that increasing human demonstration data can significantly improve robotic dexterity, emphasizing the importance of human data over machine-generated data [4][6]. - Deep Intelligence, founded in 2025, is recognized as a pioneer in leveraging human-first perspective data to decode physical common sense, which is crucial for the advancement of embodied intelligence [11][8]. Group 2: Understanding Physical Common Sense - The article highlights the critical role of physical common sense in achieving true robotic intelligence, with Generalist AI labeling it as the "dark matter" of robotics [8][14]. - Current domestic discussions in embodied intelligence often overlook the significance of physical common sense, focusing instead on trajectory fitting from real or simulated data [17][18]. - Deep Intelligence's approach prioritizes "understanding first, action next," aiming to equip robots with a deep understanding of physical world operations before executing tasks [20][21]. Group 3: Technological Innovations - Deep Intelligence has developed a comprehensive technology stack that includes data, architecture, and algorithms to enhance the efficiency of learning from human-first perspective data [24][22]. - The company has created a translation pipeline, Egocentric2Embodiment, to convert human perspective videos into structured learning signals for robots, ensuring they understand the underlying physical interactions [25][34]. - The PhysBrain model, trained on human-first perspective data, has achieved a 67.4% success rate in tasks, outperforming competitors that relied on extensive machine trajectory data [27][29]. Group 4: Advanced Model Architectures - The TwinBrainVLA architecture allows for the simultaneous training of a "left brain" for general understanding and a "right brain" for specific robotic actions, preventing knowledge loss during optimization [31][32]. - The integration of various innovations has led to the development of PhysBrain1.0, which achieved a remarkable 79.8% success rate in testing, surpassing industry benchmarks [37][38]. - The model's ability to generalize across different tasks and platforms indicates a significant advancement in robotic intelligence, showcasing the potential for real-world applications [39][40]. Group 5: Future Directions - Deep Intelligence plans to scale human-first perspective data collection to a million hours by mid-2026, aiming to fully reveal the scaling laws of physical common sense [43][46]. - The company’s approach is expected to create a competitive edge that is difficult for others to replicate, as it emphasizes data efficiency and systematic modeling of physical common sense [42][46].

Group 1 - The article discusses the "HALO trading" phenomenon, indicating that the market is beginning to anticipate changes in industry organization due to AI, with potential downward pressure on valuation centers in sectors that may be replaced by AI or where excess profits could be compressed [6] - Short-term market characteristics show that A-shares have reacted weakly to long-term tech narratives post-Spring Festival, while responding positively to current "new and old economy inflation," influenced by the "HALO trading" reflection in A-shares and the impact of Federal Reserve easing expectations [6] - The main source of short-term inflation direction is seen in cyclical commodities like steel and coal, which have recently surged, but the sustainability of these price increases is uncertain as demand verification is expected in March-April [6] Group 2 - A-share valuations as of February 27, 2026, show the CSI All Share (excluding ST) PE at 22.8x and PB at 1.9x, positioned at the 83rd and 53rd historical percentiles respectively [8] - The Shanghai Stock Exchange 50 has a PE of 11.5x and PB of 1.3x, at the 58th and 37th historical percentiles, while the CSI 300 has a PE of 14.1x and PB of 1.5x, at the 64th and 38th percentiles [8] - Industries with PE valuations above the 85th percentile historically include real estate, automation equipment, retail, electronics (semiconductors), and IT services/software development [8] Group 3 - The article highlights the emergence of AI-driven price increases in certain sectors, with a focus on glass fiber and optical fiber as investment opportunities due to visible price increases and favorable valuations [13] - Quantum technology advancements include the successful manufacturing of optical quantum chips at wafer-level high yield by a Peking University team, indicating progress towards commercial applications in quantum networks [10] - The article notes that the performance of commodities is stable during periods of PPI increases, with energy and industrial metals showing significant average gains, while stock market performance is influenced by underlying drivers such as global liquidity conditions [16]

Investment Rating - The report assigns an "Overweight" rating to the robotics industry [27]. Core Insights - The latest embodiment intelligence model, Pi 06, has significantly improved robustness, with the number of domestic humanoid startups valued at over 10 billion increasing to six [2][4]. - Pi 06 achieved a 92% autonomous operation rate in laundry tasks and a packaging efficiency of 165 items per hour [4][6]. - Nvidia's DreamDojo model has addressed the bottleneck of insufficient operational training data for robots by utilizing a large dataset of 44,000 hours of first-person human video [4][6]. - Major financing events include AI² Robotics raising over 10 billion RMB, Qianxun Intelligent completing nearly 2 billion RMB in two rounds, and Yinshi Robotics securing several hundred million RMB [4][26]. Summary by Sections Industry News and Company Dynamics - Physical Intelligence's Pi 06 model demonstrated a 92% autonomous operation rate in a laundry setting and a packaging efficiency of 165 items per hour [6]. - Nvidia's DreamDojo model has created a comprehensive robot world model, bridging the embodiment gap through large-scale human video pre-training [6]. - Domestic companies like Yinshi Robotics and AI² Robotics are making significant strides in funding and technology development [6][26]. Investment Recommendations - Focus on robotics manufacturers and core component suppliers, including: 1. Actuators and motors: Recommended companies include Zhao Wei Electric and related firms like Weichuang Electric and Buke Co [27]. 2. Reducers: Key players include Lide Harmony and Blue Dai Technology [27]. 3. Screw components: Recommended company is Hengli Hydraulic [27]. 4. Equipment: Recommended company is Tsunami Machine Tool China [27]. 5. Sensors: Recommended companies include Anpeilong and Hanwei Technology [27]. 6. Joints: Recommended company is Changying Precision [27]. 7. Structural components: Related company is Ningbo Huaxiang [27]. 8. Complete machines: Recommended companies include UBTECH and Hangcha Group [27].