At this stage, all kinds of judgments from the market and industry are pretty chaotic. For example, a professional AI blogger I follow has started to believe that Deepseek is very powerful and could replace closed-source AI. Some people even think that under the current technical trajectory, AI's 'intelligence' has already hit its ceiling. Even the guys who just participated in equipment factory audits and production line research are starting to waver. These upstream hardware companies are projecting full production and sales for 2027-2028, but what about 2029? What about 2030? At times like this, we need breakthroughs from model factories, application companies, or industries like pharmaceuticals and materials that use AI for R&D. Combine that with international dynamics and interest rate adjustments, and only then can we see the second wave of a bull market. Until then, just focus on actively doing T+0 to cut costs and quietly hold on.