OpenAI just dropped Sora 2, and it's the first time we're seeing synchronized audio-video generation
Every movement flows incredibly smoothly
And here's the kicker: it lets real people jump into scenes, so anyone can remix together
Long story short:
1 What's new: Sora 2 model plus the Sora iOS social app. This is basically the "GPT-3.5 moment" for video generation
2 Core breakthrough: First-ever synchronized audio-video generation, massively improved physical realism, and coherent narrative videos up to 16 seconds long
3 The surprise feature: "Cameo" functionality lets users blend real-person footage into AI scenes. "Remix" enables community collaboration and iterations
4 Release timeline: October 1st, 2025, launching first in the US and Canada, then rolling out globally
5 Access codes: Both the iOS app and web version at sora.com require invitation codes
The API launches in the coming weeks, while the Android version is still in development
Model Showcase
Prompt: Two mountain explorers in bright technical shells, ice crusted faces, eyes narrowed with urgency shout in the snow, one at a time Prompt: Vikings Go To War — North Sea Launch (10.0s, Winter cool daylight / early medieval)... Prompt: in the style of a japanese anime, a jubilant scene under a limitless bright sky brimming with clouds and sparkling stars... Prompt: in the style of a studio ghibli anime, a boy and his dog run up a grassy scenic mountain with gorgeous clouds, overlooking a village in the distant background Prompt: man does a cannonball off diving board Prompt: a man rides a horse which is on another horseThe Sora App
Along with this update comes an iOS app
Search for "Sora by OpenAI"
Plus there's a web app
Visit: https://sora.chatgpt.com/
As for features, it includes standard text-to-video and image-to-video generation. According to OpenAI:
Start from a prompt or uploaded image to create videos in any style with unprecedented realism: cinematic, animated, photorealistic, or surreal
You can also star alongside your friends
Through the Cameo feature, you and your friends can become characters in the video, breaking down any dimensional barriers
Remix Everything
Freely swap characters, change the mood, add new scenes, or continue the story. Like letting cats rule everything
But don't get too excited...
All of this requires invitation codes
Where did they learn this gatekeeping strategy from?
Timeline and Other Details
For context on what came before and after, I created this timeline to help clarify things
Final Thoughts
OpenAI really has something here
