In the video below, OPenAI’s CTO, Mira Murati discusses with WSJ’s Joanna Stern how their new text-to-video AI model Sora works.
How long does it take to produce a result? As it’s a technology, that is still in its infancy, what about accuracy and consistency and managing the inevitable glitches and imperfections? When are we going to be able to edit the results, in post-production, as it were? And, as you will see from the video, there are unanswered, shall we say carefully fielded, questions around the data that was used to train the Sora machine learning model.
To talk about how to safely and reliably use generative video and other machine learning-based content origination tools, please get in touch. In the first instance, your Mondatum contacts are Colin Birch (colin@mondatum.com) and John Rowe (john@mondatum.com).
Source: Wall Street Journal