Open ChatGPT. Ask it: "Where should I take my kid for a weekend in Shanghai?" It will hand you a clean, well-structured list. Locations, transit, restaurants, things to watch out for, even a backup plan if it rains. It reads like advice from a local friend who knows the ropes. But have you ever stopped to ask—where does it know all this from? Not Disney's official site. Not Ctrip. Not Dianping. It's Xiaohongshu. It's a note some mom posted two years ago, with weather tips, parking advice, and the pitfalls she stepped into on the ground. The model never tells you who it read. But its answers have a texture. And that texture is the texture of UGC.
This is a watershed moment. OpenAI's full rollout of its advertising platform to enterprises worldwide is more than a commercial milestone — it marks a deep fracture in the AI industry's trajectory: AI products are now splitting into two distinct paths, attention monetization and productivity monetization.
If you opened ChatGPT today and asked, "I'm considering working with Company X — are they trustworthy?" what would you get back? Most likely, a measured, balanced response citing multiple sources. It sounds fair. But buried inside might be a sentence like this: "The company faced media scrutiny several years ago over a controversy that raised consumer concerns." To the AI, that's just a retrieved factual statement. To the brand, that controversy may have long since blown over — officially clarified, internally addressed, and put to rest.
OpenAI has long pointed out in multiple safety and research reports that generative AI systems are vulnerable to data poisoning and prompt injection. At their core, these risks define the very challenge GEO (Generative Engine Optimization) now faces: How can a brand be correctly constructed and trusted within AI systems’ mechanisms of understanding, citation, and recommendation?
