Generative AI has been making progress in realism, particularly in text-to-image tools. However, text-to-video, or AI-generated videos, have been less realistic. OpenAI's Sora is set to revolutionize this by creating one-minute-long videos based on text prompts, which are very realistic. These videos include human facial elements and ambient scenes. The technical report reveals that the inspiration comes from large language models and generalist capabilities, including recurrent networks, generative adversarial networks, autoregressive transformers, and diffusion models. The Sora can handle complex scenes, handle multiple subjects and elements in the same frame, reproduce a sense of motion, and return realism with details. This makes it more difficult to distinguish AI videos from real-world videos. The more detailed the text prompts, the richer detailing models will generate, making it more difficult to distinguish AI videos from real-world footage.