OpenAI Makes a Bold Move into Hollywood with New Video Creation Tool
San Francisco-based start-up OpenAI has made a significant foray into the world of Hollywood with the launch of its latest tool – a generative artificial intelligence system capable of producing highly realistic videos. This marks OpenAI’s first major step into the entertainment industry and has the potential to disrupt traditional labor in areas such as visual effects, sound engineering, voice acting, and concept art.
The new system, called Sora, has the ability to generate short videos of complex scenes with multiple characters, various types of shots, and mostly accurate details. The videos are created in response to a text prompt and can be produced in minutes. OpenAI showcased its capabilities by unveiling videos including a movie trailer featuring an astronaut on a desert planet and an animated scene of a creature next to a melting candle.
While AI video tools have been available before, OpenAI’s system stands out for its potential to revolutionize the industry. A recent survey conducted among 300 Hollywood leaders revealed that three-fourths of respondents believed that AI tools would lead to job elimination, reduction, or consolidation within their companies. It is estimated that approximately 204,000 positions will be affected within the next three years.
Professionals such as sound engineers, voice actors, and concept artists are at the forefront of this potential displacement. Visual effects and postproduction work are also perceived as particularly vulnerable. The arrival of OpenAI’s tool has sent shockwaves through the industry, triggering concerns about the future of these roles and prompting many to consider alternative career options.
OpenAI’s Sora is unique compared to other AI video generators as it can create longer videos while maintaining visual quality and consistency, adhering to the users’ prompts. It allows for the seamless switching of shots with different compositions, including close-ups, tracking shots, and aerial views. However, there are still some limitations. In one example, a video of a woman walking down a Tokyo street showcased a flaw where her legs switched sides midway through the clip. OpenAI acknowledged that the current model has weaknesses and may struggle with accurately simulating complex scenes and understanding cause and effect.
In an effort to address potential issues, OpenAI is undertaking safety testing before releasing Sora to the public. The company has invited experts in misinformation, hateful content, bias, visual art, design, and filmmaking to provide feedback and suggestions for improvements. Additionally, OpenAI plans to embed metadata in videos produced by the system to demonstrate that they were created by AI. They will also reject prompts that violate usage policies, including extreme violence, sexual content, celebrity likenesses, or intellectual property infringement.
OpenAI has not disclosed the materials used to train its system, no longer revealing the source of training data to maintain a competitive advantage. However, the company is currently facing lawsuits from several authors who claim that OpenAI used their copyrighted books without permission. If these lawsuits are successful, OpenAI may be required to destroy systems that were trained on copyrighted material.
The rise of AI video tools poses a threat to commercial production, where quality may be considered less important compared to film and TV production. The decline in commercial shooting in Los Angeles since 2018, reaching a seven-year low in 2023, further emphasizes the potential impact of AI tools on the industry.
Overall, OpenAI’s move into Hollywood with its new video creation tool has raised concerns about job displacement and the future of the entertainment industry. While the system shows promise, there are still limitations to overcome. The industry must adapt to these rapid advancements in AI technology, finding new ways to harness its potential while ensuring the preservation of creative human talent.
