My recommendation of Micron at $120 last August, based on today's perspective, could be considered a recommendation made on the eve of a storage explosion. If agents like Clawdbot become widespread, AI will generate data a million times faster than humans! The only limiting factor is electricity. However, according to Musk's vision for SpaceX's space-based solar data centers, once Starships mass-produce computing power and power equipment to space, this problem will be quickly solved. The time window is 28-30 years to reach a terrifying TW-level scale. In space orbit, solar panels are unaffected by clouds and the atmosphere, and there is no day-night cycle (when in a twilight orbit), resulting in power generation efficiency 7-10 times that on the ground. Based on SpaceX's space plan, assuming the US has no power shortage for 28-30 years, storage demand will continue to grow exponentially. For now, to ensure stability, I've chosen Hynix. Furthermore, with the widespread adoption of user-side agents, user-level storage demand will shift to AI servers, and the customer-level storage market will shrink rapidly! I'd guess this differentiation will become apparent quickly once agents like Clawdbot become widely available, causing user-level storage prices to return to normal levels. Meanwhile, HBM3 storage, needed by AI servers, will remain at a high price due to extreme scarcity.

I can copy homework again! 🥰

I saw that, but missed the key to wealth.

Good targets are harder to find this time of year compared to last year. They've been thoroughly explored and hyped up by the market.

Get rich

Today I was thinking of making a killing on the Sino-Korean semiconductor market, hoping to get in after yesterday's crash 😂. I placed my order at 3.1, hoping to catch a falling knife, but it opened at the daily limit up. LOL.