Recent advancements in image generation technology are leading to significant improvements in quality and reliability. Traditionally generated images, especially for low-budget ventures like YouTube channels, often include nonsensical details, such as distorted text and odd combinations of furniture. However, innovations are promising to eliminate these flaws, enhancing the visual output of generative AI tools. This raises questions concerning user engagement, as many users express a lack of excitement compared to expectations from previous years. The new approach to generating images introduces a method akin to how large language models (LLMs) create text, generating visuals one 'token' at a time. This could inspire further curiosity into how such image tokens can be interpreted and utilized. Additionally, the meme generation capabilities of AI are advancing, hinting at a stronger cultural integration of humor within this technology, a key aspect of creative expression.