The demand for human-generated data is intensifying in AI development, with concerns that a data shortfall could occur by 2028 as companies exhaust readily available datasets. In response, companies are increasingly seeking proprietary data sources found within financial and academic institutions, highlighting that a lack of free data does not equate to a complete data shortage. Additionally, there is a need to transform various forms of media, such as television and videos, into usable data formats, driving advancements like OpenAI's Whisper. The trend towards multimodal models—capable of processing language, images, and video—presents challenges, as current models may struggle to manage this complexity effectively. Despite some skepticism regarding the scalability of these models, insiders indicate that innovative strategies are emerging to navigate the anticipated data wall, suggesting that progress continues in scaling AI capabilities.