AI Film News | February 16, 2024
Here’s the latest news in the world of AI Filmmaking.
AI Film News | February 16th, 2024
In this video, we talk about some of the latest new technology in AI Filmmaking and what you need to know. We also highlight some incredible AI Filmmaking projects. You can find the links shown in the video below.
Join Our Newsletter for the Latest AI Filmmaking News.
Check out our AI Filmmaking Course.
Here is a breakdown of the news from the video.
OpenAI Announces Sora - Open AI announced their brand new text-to-video tool, Sora. The tool creates videos that are almost indistinguishable from reality. It's the biggest step forward in AI video since Runway Gen-2.
Adjust Emotion Sliders - EmoSpeaker is a new research paper that allows you to adjust actor performances with a slider. It's very uncanny now, but will likely be within editing tools in the next 2 years.
Better Motion Control with Boxes - This new research tool allows you to have more control over movements by defining the beginning and end of the animated paths.
Leonardo Releases Photoreal 2 Model - Leonardo created a new image model for creating photorealistic results. The outputs are about as good as Midjourney 5.2.
New Image Upscalers - Both Krea and Leonardo released their new image upscalers this last week. We actually did a game in the comments of our Film News video to get $500 if you guess them correctly.
Gemini vs ChatGPT - We tested Google Gemini for creativity this last week and were impressed by the results. Obviously there are limitations, but currently, Google Gemini gives you better film loglines and creative ideas.
ChatGPT has a Memory - ChatGPT will now be able to remember details between conversations to keep you from having to train it again and again when you create a new chat.
Apple Dabbles in AI Motion Design - Apple is working on a brand new Motion Design tool that can automatically animate shapes.
A Great Twitter Follow - Hugo Ventura is a fantastic resource on Twitter for learning AI prompting & techniques. We really liked some of his contextual experiments.
NVIDIA Releases Local Chatbot - NVIDIA announced a new chatbot that can run entirely locally on your machine. It's about as good as ChatGPT 3.5 but completely offline.
Elevenlabs Multilingual Speech-to-Speech - Elevenlabs released Speech-to-Speech in 29 different languages this last week. Paving the way for automatic dubbing in virtually every language the near future.
Become an ElevenLabs Voice - ElevenLabs will now pay youevery time your voice is used on their platform.
Consistency with Single Prompt - A team of researchers announced the ability to get consistent characters in all of your generations from a single prompt using SDXL. It also can generate about 10 images per second!
Superbowl AI Confusion - Some folks were very confused that the Jesus ads in the Superbowl weren't AI generated. The photographer came out and said it wasn't AI… but it was extremely photoshopped so it was hard to tell.
