AI infrastructure, I think, is currently at a stage of high valuation, or even considered mid-term in the long run, having overdrawn a lot of future growth potential. Even a toilet manufacturer has surged due to AI, which shows how crazy it is, but that doesn't mean there's no room for hype.

After intense training and building infrastructure for AI, it ultimately comes down to AI implementation. Both virtual and real-world applications require AI deployment to recoup the massive infrastructure costs, but virtual deployment is the fastest.
AI will reshape or reconstruct all industries, including the virtual world. Therefore, some software companies, as representatives of virtual productivity, are the easiest to implement AI in, and they are currently severely undervalued and misjudged.
In the later stages of hype in the AI infrastructure sector, I will gradually cut into positions. From this round of corrections, they have performed quite well, but when funds will rotate into them remains unknown.
Buying when no one is paying attention also means enduring patience. Investing is like that, and speculation is my stubborn act of going all-in with conviction.