Meitu / 美图
Meitu is the central company case in 263.Sora死了，Adobe跌了，美图何去何从？. The episode presents Meitu as moving from consumer beauty tools and past platform ambition toward AI application work across production tools, globalization, ecommerce advertising, and agent-callable capabilities.
The source’s Meitu thesis is that a mature tool company can survive model pressure if it keeps moving: raw model output may absorb generic features, but Meitu can still own AI Application Layer Moat through beauty-specific detail, visual taste, user behavior data, ecommerce material workflows, style consistency, batch generation, ROI feedback, and Model Container Strategy. 吴欣鸿’s “AI team” framing makes the company less a single editor and more a workflow layer that coordinates multiple capabilities for a result.
Key Points
- Meitu XiuXiu / 美图秀秀, BeautyCam / 美颜相机, Meipai / 美拍, Meitu Design Studio / 美图设计室, Kaipai / 开拍, and Wink anchor the episode’s Meitu product history.
- The source says Meitu’s production-tool and globalization push aims to raise paid conversion and ARPU after its user base is relatively stable.
- Meitu Design Studio / 美图设计室 is framed as a vertical workflow product for ecommerce, advertising, and design material generation.
- Kaipai / 开拍 began from a teleprompter-like function in BeautyCam and became a separate product after usage data showed demand.
- Meitu’s competition with Jianying / 剪映 is not treated as a winner-take-all video-editing fight; the source emphasizes narrower scenarios where Meitu has stronger beauty or visual-detail know-how.
- Meitu’s model posture is Model Container Strategy, using external APIs, open-source models, self-trained vertical models, and product-layer judgment depending on the task.
- Meitu packaging capabilities as AI Skills is treated as a possible entry into To-Agent Distribution.
Connections
- 吴欣鸿, 魏熙, and 庄明浩 / 庄明昊 — people framing the Meitu case in the source.
- AI Application Layer Moat, Vertical Workflow AI, and Model Container Strategy — core strategy concepts.
- Adobe, Sora, and Model Provider Tool Competition — contrast cases for application-layer pressure.
- AI Visual Merchandising, Product Led Willingness To Pay, and Domain Expert Alignment — existing concepts extended by the source.