The wave of AI applications has indeed arrived, but I think it’s important to view it calmly.

Recently, the hot topic is the interaction between AI and the Spring Festival Gala. Such interactive projects can indeed generate some hype from an emotional perspective. The problem is, from a fundamental standpoint, this is similar to the development logic of some popular applications last year — the theme is hot, but performance support still needs time. If we have to compare intensity, this wave may not necessarily be more explosive than last year’s AI breakthrough events.

Why do I say that? I have three points:

**First, the chip market is not clean.** AI applications are not a new concept; there are long-term market participants, so the capital structure is relatively complex and prone to fluctuations.

**Second, the policy environment is cautious.** Countries’ attitudes toward advancing AI are measured. On one hand, AI can indeed improve productivity; on the other hand, it impacts employment structures, which conflicts with current social policy frameworks, limiting policy space.

**Third, business models are still being explored.** This is key. The revenue status of overseas service-oriented AI platforms is actually average, and the same applies domestically. If user payments are truly required, how many people are willing? This uncertainty makes institutions cautious.

Of course, this direction will have ongoing momentum, but if you want to make big money, the target needs to be narrower, and focus should be on key areas. Following the trend blindly won’t lead to high efficiency in making money.