1. OpenAI’s Sora AI video generator can produce impressive visuals, but may not be ready to replace traditional filmmaking techniques.
2. The short film “Air Head” by the Shy Kids production studio used Sora clips, but had to rely on traditional VFX techniques to fix inconsistencies.
3. Working with Sora is a slow process, with clips taking up to 20 minutes to render and requiring detailed prompts to ensure consistency. It is likely to be used as a supplementary tool in filmmaking rather than a complete replacement.
OpenAI’s Sora AI video generator, while impressive, may not be quite ready to replace Hollywood just yet. A short film called Air Head by Canadian production studio Shy Kids showcased the use of Sora, but it was revealed that traditional VFX techniques had to be used to fix consistency issues with the generated shots. This lack of transparency may not show OpenAI in the best light.
Access to Sora has been limited due to the cost and time of generating clips, as well as safety concerns surrounding potential use cases for high-quality synthetic content. The process of working with Sora is not fast, with clips taking up to 20 minutes to render, and adjustments needing to be made to avoid a slow-motion effect.
While working with Sora, creative professionals like those at Shy Kids have had to approach the process as more of a documentary than a traditional film, working with a massive amount of generated material and crafting it into a story. The use of AI-generated content in filmmaking is likely to be part of a workflow rather than a replacement for traditional filmmaking.
Despite the challenges and limitations of using Sora, there is potential for AI video to lead to a new generation of films and possibly a new golden age of cinema. Adobe has also integrated generative video into Premiere Pro, indicating how AI video could potentially be used in the industry. Overall, while there are still hurdles to overcome, the future of AI video in filmmaking is promising.