OpenAI's CTO Discusses Sora's Breakthroughs with WSJ

OpenAI’s new text-to-video AI model Sora can create some very realistic scenes. How does this generative AI tech work? Why does it mess up sometimes? When will it be released to the public? What data was it trained on?
WSJ’s Joanna Stern sat down with OpenAI CTO Mira Murati to find out.
Chapters: 0:00 OpenAI’s Sora 1:08 Reviewing Sora’s videos 4:03 Optimizing and training Sora 6:42 Concerns around Sora 8:55 Identifying Sora videos


I liked that she asked about generation time. DW: It can take a few minutes depending on the complexity of the prompt.

A few minutes itsn’t that bad. But think about this. in a year or so that will be 15 to 30 seconds. You will be able to host a DnD game or any board game and the dungeon master or host can start generating the videos to set the scen based on what the players are doing. Less expensive commercials, facebook ads will get crazy, animators wont know what to do, this is going to change everything.


Welcome to the community!

Sorry if my post was a bit confusing. I was saying that it was a good thing that it was only a few minutes, as some other platforms take a few minutes for 10-15 second videos.