Core Viewpoint - The discourse surrounding large models consuming applications has raised market concerns, leading to a collective setback in the AI application sector. Despite the release of Nano Banana, Meitu's application data continues to grow rapidly, indicating a synergistic effect between general large models and applications [1][3]. Group 1: Insights from CEO Wu Xinhong - Wu Xinhong, CEO of Meitu, acknowledges that while general large models are "omnipotent," the space left for application layers is diminishing. However, the efficiency of general large models in specific vertical scenarios is not very high [1][3]. - He compares large models to a "Swiss Army knife," capable of handling general needs and daily tasks, while vertical applications are likened to specialized tools that meet specific demands in various scenarios [1][3]. - Wu emphasizes that application developers always have opportunities at different stages, with the key being the deep exploration of high-value vertical scenarios that exhibit rigid demand and high costs, where customers are willing to pay, thus creating high elastic growth potential [1][3]. Group 2: Limitations and Focus Areas - Wu believes that the conversational interaction of current general large models has limitations, and the threshold for extracting vertical industry capabilities is high, necessitating vertical applications to unleash the potential of large models [2][4]. - Specific vertical application scenarios such as industry SOPs, vertical creator communities, high-precision editors, consistent batch outputs, material asset management, and team collaboration are areas where general large models may not perform well [2][4]. - Meitu is committed to becoming a platform that continuously generates high-quality imaging applications, focusing on creating more vertical scenario imaging products [2][4].
美图吴欣鸿回应大模型竞争：垂直应用好比专业工具 美图应用数据仍快速增长