Christian, founder of Infini: The anthropomorphic expression of big models has become the core threshold, and Agent moats are shifting from models to application layers BlockBeats News: On April 21st, during a roundtable discussion on "Cracking Web 4.0: When AI Agents Take Over Chain Permissions," Christian, the founder of Infini, stated that from a product and growth perspective, the core competitiveness of AI still depends first and foremost on the language processing ability of large models, especially the "personification degree" of content output. He pointed out that compared to early models that tended towards mechanized and template based response methods, current users are more concerned about whether AI can provide expressions and content quality that are close to real human conversations, which directly affects its practical value in external communication and team collaboration. Christian mentioned that as model capabilities continue to evolve, basic information processing capabilities are becoming increasingly homogeneous. When large models can complete about 90% of standardized tasks, relying solely on the model itself is no longer a long-term barrier. The competition among AI agents is gradually shifting from model capabilities to specific application scenarios and workflow design. In the future, the ability to truly differentiate will no longer be limited to simple information processing or desktop automation, but will be achieved in complex scenarios such as finance and trading through a deep understanding of business processes, enabling higher-level task execution and decision support.