OpenAI and NVDA are both at the point of making strategic decisions. The root cause is that NVDA took away most of the FCF in the entire chain (by the year 27), causing others to fight for survival. From the development trajectory, it is necessary to invest in OpenAI as soon as possible to obtain AI tickets Secondly, the entire top AI laboratory may only have those 50 talents? All three companies have split up (Open/Google/Anthopic), and what's even more frightening is that these three companies can all be seen as talents cultivated by Google over the past decade, who are more familiar with Google's system and style. Therefore, if NVDA wants to seize the opportunity from the source, it can only make a bet on Open from the split companies, and I feel that a significant investment of billions is needed The big difference between AI and the Internet lies in the transmission cycle. It may take two years for AI to basically complete the whole process, from papers to laboratory internal effects to certain application presentations, to workshops to supply chains to performance. Greg Brockman of OpenAI once said that they had hysterical efforts to change the architecture and chip design early, but no one heard of it. We are too weak, and it is too difficult to change the design of the whole supply chain, which leads to cognitive transmission failure. Therefore, if you want to keep the moat, you can only start from the very source, just like acquiring Groq, directly locking the reasoning end SRAM in your hands So the AMD INTC opportunity has come because the demand is too high