OpenAI Sora is a new AI model that can create one-minute videos from text prompts and it is mindblowing

OpenAI says that Sora can produce intricate scenes featuring multiple characters, precise movements, and detailed backgrounds. The model not only comprehends the user’s prompts but also interprets how these elements manifest in real-world scenarios. “The model has a deep understanding of language, enabling it to accurately interpret prompts and generate compelling characters that express vibrant emotions. Sora can also create multiple shots within a single generated video that accurately persist characters and visual style,” says OpenAI.

While the OpenAI Sora model has added a lot of excitement to the internet this morning, popular YouTuber Marques Brownlee aka MKBHD has rightly pointed out the possible concerns around the model. “Every single one of these videos is AI-generated, and if this doesn’t concern you at least a little bit, nothing will,” he wrote in his post.

The capability of the OpenAI Sora model undoubtedly appears impressive at the moment. However, being wary of a model that can effortlessly generate a one-minute video from simple text prompts is crucial, as it could be easily misused. If you recall the AI-generated video of Will Smith eating spaghetti that went viral in early 2023, you’ll recognize the significant leap that Sora represents from that. There was also another one featuring Scarlett Johansson eating something. Sora demonstrates a massive improvement with its uncannily realistic-looking videos.

However, OpenAI is not shying away from a conversation around safety. The company says before releasing Sora in OpenAI’s products, they will be implementing several crucial safety measures. This includes collaborating with red teamers, experts in areas such as misinformation, hateful content, and bias, who will rigorously test the model to identify potential vulnerabilities. Additionally, OpenAI will be developing tools to detect misleading content, such as a detection classifier capable of identifying videos generated by Sora.

OpenAI will also be leveraging existing safety protocols developed for its products that use DALL·E 3, which are relevant to Sora as well. For instance, within an OpenAI product, OpenAI’s text classifier will screen and reject input prompts that violate usage policies, such as those requesting extreme violence, sexual content, hateful imagery, or celebrity likeness. The company say it has also established robust image classifiers to review every frame of generated videos, ensuring adherence to our usage policies before user access.

OpenAI also says that it actively engaging policymakers, educators, and artists globally to address concerns and explore positive applications of this new technology. “We’ll be engaging policymakers, educators and artists around the world to understand their concerns and to identify positive use cases for this new technology. Despite extensive research and testing, we cannot predict all of the beneficial ways people will use our technology, nor all the ways people will abuse it. That’s why we believe that learning from real-world use is a critical component of creating and releasing increasingly safe AI systems over time,” says OpenAI.

Currently, Sora is only available for the red teamers to check out the important areas for potential issues or risks. OpenAI is also opening up access to visual artists, designers, and filmmakers so that the company can gather input on how to improve the model. “We’re sharing our research progress early to start working with and getting feedback from people outside of OpenAI and to give the public a sense of what AI capabilities are on the horizon,” reads the blog.