1. Froth is localized. Hardware (optical modules, computing power) has increased significantly, and there is indeed some hype; but applications (AI programming, customer service, live streaming) are still in the early stages, and the truly profitable direction is applications, not selling shovels.

2. Nvidia's earnings report (after market close on May 20) is a watershed. Expectations are already maxed out; just being "passable" isn't enough, it needs to "beat expectations + guidance explosion" to continue rising. Otherwise, the hardware sector will take a hit.
3. Both A-shares and U.S. stocks are doing the same thing: funds are shifting from hardware to applications. Policies are also pushing (computing power into new infrastructure), ByteDance and Alibaba are still investing heavily, but hardware stocks have already started to fall. Don't chase high on hardware; wait for a pullback to buy applications.

The only risk: U.S. Treasury yields are too high (the 10-year is nearly 4.6%), and tech stock valuations are under pressure. If Nvidia underperforms, everything will follow.

So my strategy: don't touch hardware for now, wait for Nvidia's earnings to be confirmed. On the application side, wait for a pullback to buy low, don't chase the rally.

Do you think Nvidia can beat expectations this time? Drop your bets in the comments. $BTC ‌#Polymarket每日热点