瑞银：中国出现AI泡沫的概率不高，变现靠云与广告

Group 1 - The emergence of DeepSeek in early 2025 has significantly increased global attention on China's AI sector, prompting many overseas investors to reconsider Chinese assets, particularly in the technology sector [1] - UBS analysts believe that by 2026, the development paths of AI in China and the US will diverge, making foreign investment in Chinese assets a balanced risk choice [1][9] - Compared to the US, the probability of a systemic AI bubble in China is much lower, as leading Chinese model firms rely more on cash flow from parent companies rather than excessive financing [9] Group 2 - The primary monetization models for AI in both China and the US are cloud services and advertising, with cloud revenue expectations being continuously revised upwards [2] - In the consumer market (ToC), the progress of AI monetization in China is slower than in the US due to factors like product maturity and user willingness to pay [2] - By 2026, consumer payment for AI products in China may accelerate only when these products provide clear, repeatable, and quantifiable value [2] Group 3 - China is expected to focus more on vertical applications of AI, while the US is betting on general artificial intelligence [3] - Despite potentially lower revenue contributions from AI in China compared to the US, the return on investment may not be at a disadvantage due to fewer constraints in infrastructure [3] - AI is currently being used to transform existing businesses, with gaming and advertising being the most prominent application areas [3] Group 4 - The evolution of AI agents is seen as a phased process, facing challenges in technology, user acceptance, and regulatory frameworks, which may delay large-scale deployment and monetization [4] Group 5 - Chinese companies are adopting AI more proactively than their US counterparts, integrating AI functionalities into their native products, which reduces the risk of disruption from upstream AI developments [10][11] - Major Chinese internet firms are viewed as the biggest beneficiaries of AI, with additional interest in brokerage firms and the solar energy sector due to attractive valuations and global energy trends [11]