东吴汽车黄细里团队·2025-06-29 12:08Core Viewpoint - The article emphasizes the importance of automotive intelligence as a revolutionary shift in transportation, predicting that 2025 will be a pivotal year for the industry, with significant increases in L3 automation penetration rates and a competitive landscape that favors strong players [2][4]. Group 1: Automotive Intelligence Revolution - Automotive intelligence is described as a transportation revolution, with three key components: L3 automation aiding car sales, L4 Robotaxi enabling software monetization, and the global rise of domestic brands [2][4]. - The penetration rate of L3 automation is projected to increase from 10% in 2025 to over 80% by 2027, driven by consumer demand and new vehicle cycles [2][4]. Group 2: June Summary of Intelligence - In June, the penetration rate of urban NOA (L3 automation) reached 22.0%, marking the first time it surpassed 20% [3][9]. - The overall urban NOA penetration rate for the Li Auto 2025 model reached 54.9%, a month-on-month increase of 10.6 percentage points, indicating a significant shift in consumer demand [3][9]. Group 3: Key New Models - The Xiaopeng G7 Ultra was launched with the industry's first L3-level computing platform, featuring three Turing AI chips with an effective computing power of over 2200 TOPS [3][9]. - The Xiaomi YU7 model includes the NVIDIA AGX Thor chip, enhancing its performance with a threefold increase in clip count compared to the previous version [3][9]. Group 4: Investment Recommendations for 2025 - The article recommends focusing on intelligent vehicles as the core investment theme, with a preference for Hong Kong stocks over A-shares, highlighting companies like Xiaopeng Motors, Li Auto, and Xiaomi Group [4][8]. - It also suggests investing in incremental components related to AI chips and domain controllers, naming companies such as Horizon Robotics and Desay SV [4][8]. Group 5: Market Outlook and Trends - The automotive intelligence market is expected to see a significant increase in the penetration of new energy vehicles, with projections indicating a rise to 50%-80% by 2027 [18]. - The article outlines a two-phase consumer willingness to pay for intelligence, with the first phase focusing on hardware sales and the second phase aiming for software monetization [18].