On Feb. 15, OpenAI introduced the world to Sora, a new AI model that can generate more than a minute long video from a text prompt. According to OpenAI and with some examples, we were able to see the capacity of this innovative technology and it is more than impressive. All the user has to do is type out exactly what he or she wants and the AI model generates it.
Computer-generated videos are not new to the public. We have seen cases of deep fakes, but these videos are created by teaching an AI to learn certain things such as the facial expressions and speech movements from an already existing video, the source material, and then applying those same patterns of movement unto another person’s face. With Sora, however, all the user has to do is type exactly what the user wants, and it creates it.
At first, there were some doubts on the internet but Sam Altman, the CEO of OpenAI, the company that created ChatGPT and now Sora, made an open request for new prompts from people on X, the former Twitter. One user, @fxturevescent, requested a video of two golden retrievers podcasting on top of a mountain. In about an hour Altman replied with a video of two golden retrievers podcasting on a mountain. One of the amazing things about this technology is that this is just the beginning, and the videos are almost indistinguishable from real life to the human eye.
One can only imagine what the future will hold but what we can be sure of is that it will get better as time goes on. Some of the prompts used included “A movie trailer featuring the adventures of a 30-year-old spaceman wearing a red wool knitted motorcycle helmet, blue sky, salt desert, cinematic style, shot on 35mm film, vivid colors,” “An adorable happy otter confidently stands on a surfboard wearing a yellow life jacket, riding along turquoise tropical waters near lush tropical islands, 3D digital render art style.”
The hyperrealism of the videos derived from each of these prompts makes you wonder about the job security of people in the field of entertainments in the future. If this is just the beginning and we can generate movie trailers and animated videos using just a text prompt, you can see a future where movie studios would not need to have large productions in order to make anything again, just a few people feeding prompts to a machine so as to create contents for their audience. Not only that, but there were also concerns about people using this technology to generate adult content using the likeness of unsuspecting and innocent people.
Although the future with this technology looks uncanny, the truth is that it is inevitable. And just like every innovation, this technology would also bring about new jobs to people. Animation studios would need people who are good at inputting prompts, scriptwriters, and animators who would be able to retouch the generated videos. Even the public would benefit from this. A student with no funding would be able to create a short film with the help of Sora from his or her bedroom. All they need is a series of prompts about what they envision, and Sora would bring their imaginations to life.
The light bulb, printing press, airplane, telephone, the internet, AI, all these are time-stamps or chronological markers that have shown how far we have come as humans from the Stone Age. Sora is just another great invention that has come to stay.
