Is Google about to end the "uncanny valley" of AI text in video? 🧪📓 A leaked demonstration of Google's highly anticipated Gemini Omni model has just surfaced, and the results are a massive leap forward for AI video technology. In this clip, a professor writes a complex trigonometric identity on a chalkboard with near-perfect text coherence—an area where even top-tier models like Sora and Veo have historically struggled. This level of precision suggests that Google’s new multimodal models are finally mastering the fine-motor coordination and symbolic reasoning required for accurate, real-world text rendering. Technically, what we’re seeing is a breakthrough in temporal consistency and character-object interaction. Keeping the chalk, the handwriting, and the professor’s movements synchronized while maintaining a high-fidelity "traditional classroom" aesthetic is one of the most difficult challenges in AI video production. If this leak holds true for the official reveal at Google I/O ...