<Understanding X's Incentive Mechanism Through Grok's Needs> The existing data on the internet has already been mostly 'consumed' by large models. Now, the focus of competition has shifted to the 'incremental' data that search engines haven't had the chance to capture yet. This refers to the new content emerging in real-time every day: news, opinions, discussions. Although much of the new content has been polluted by AI-generated data, those authentic, professional new pieces—whether breaking news, academic papers, or literary works—still hold unique value. I think Musk has figured this out. His push for creator incentives (Post to Earn) is aimed at systematically incorporating this high-value new data into the X platform. This directly determines Grok's advantage: when you want to look up new technologies or trending events, it doesn't need to rely on external search engines; searching keywords directly within X is enough. Other AI Agents trying to achieve the same effect would either need to integrate external search engines or pay X for access to its data. But Musk only needs to share a portion of the revenue with creators to continuously obtain solid, firsthand data. Following this logic, you can probably guess what kind of content the current algorithm will prioritize.