On Thursday 15 February 2024, OpenAI’s CEO Sam Altman unveiled the artificial intelligence research organisation’s latest AI tool, Sora, which, using nothing more than a brief text prompt, can create videos up to 60 seconds long so realistic that the average person would struggle to guess they were generated by a computer. Scary, right?
AI is getting insane
— borovik (@3orovik) February 20, 2024
Will Smith eating pasta 1 year ago vs now pic.twitter.com/zRy56dCIH3
Though the video generator, which succeeds OpenAI’s infamous ChatGPT and its more advanced version GPT-4, has not been officially released yet, it’s already been made available to “red teamers to assess critical areas for harms or risks” and a “number of visual artists, designers, and filmmakers to gain feedback on how to advance the model to be most helpful for creative professionals.”
This has led to a number of mind-blowing videos made using Sora to appear online and, to be honest, some of them are so terrifyingly good that SCREENSHOT had to round up the most intriguing examples found on X (formerly Twitter):
https://t.co/qbj02M4ng8 pic.twitter.com/EvngqF2ZIX
— Sam Altman (@sama) February 15, 2024
Most of the videos shared were put out by OpenAI itself but CEO Sam Altman was also taking prompt requests such as the one above showing “a bicycle race on ocean with different animals as athletes riding the bicycles with drone camera view.” Pretty cool.
https://t.co/rPqToLo6J3 pic.twitter.com/nPPH2bP6IZ
— Sam Altman (@sama) February 15, 2024
https://t.co/rmk9zI0oqO pic.twitter.com/WanFKOzdIw
— Sam Altman (@sama) February 15, 2024
Plenty more example videos are around as seen below. Interestingly, many (if not all of them) are slowed down.
It's only been 24 hours since Open AI announced Sora, and people are going crazy over it.
— Angry Tom (@AngryTomtweets) February 16, 2024
Here are 10 wild examples you don't want to miss: (Part 2)
1. Blend individual videos togetherpic.twitter.com/XN5Xq1qGQT
OpenAI Sora is wild.
— Min Choi (@minchoi) February 16, 2024
These are 100% AI generated from text. And generate up to 1 minute! 🤯
1. PROMPT: A corgi vlogging itself in tropical Maui. pic.twitter.com/bRJBQcBH1r
So it turns out that OpenAI's Sora is even more insane than I thought. Here are some of the absolutely nutty things it can do ⬇️ pic.twitter.com/EugpRM71qt
— Joseph Mambwe (@MrMambwe) February 16, 2024
"realistic video of people relaxing at beach, then a shark jumps out of the water halfway through and surprises everyone"
— Tim Brooks (@_tim_brooks) February 17, 2024
Video generated by Sora pic.twitter.com/WRVnrguCif
This example from @OpenAI for their new text to video product blew my mind...
— Kyle Tibbitts (@KyleTibbitts) February 15, 2024
Prompt: Historical footage of California during the gold rush... pic.twitter.com/VTARGIlH4P
If you think OpenAI Sora is a creative toy like DALLE, ... think again. Sora is a data-driven physics engine. It is a simulation of many worlds, real or fantastical. The simulator learns intricate rendering, "intuitive" physics, long-horizon reasoning, and semantic grounding, all… pic.twitter.com/pRuiXhUqYR
— Jim Fan (@DrJimFan) February 15, 2024
even the sora mistakes are mesmerizing pic.twitter.com/OvPSbaa0L9
— Charlie Holtz (@charliebholtz) February 15, 2024
🚨 Good Bye Reality
— Sambhav Gupta (@sambhavgupta6) February 15, 2024
Every single video in this thread was generated by Open AI's new text to video model "Sora" 🤯
🧵
1/48 pic.twitter.com/ze4Mom2Bgs
🚨 BREAKING: OpenAI just launched Sora, an AI model that can create 60-second videos from just text prompts.
— Rowan Cheung (@rowancheung) February 15, 2024
The video below was 100% created by Sora. pic.twitter.com/qu9KN1QUMv
This video was made with the not-yet-released Sora AI technology just announced from OpenAi. This changes everything. It's 27 seconds from a text prompt.
— Ben Nash (@bennash) February 15, 2024
Here is their prompt:
Prompt: A white and orange tabby cat is seen happily darting through a dense garden, as if chasing… pic.twitter.com/8XxxFqiywC
OpenAi's Sora can do a lot more than just generate text to video. This example shows an objectively beautiful morph between one subject into another. The VFX implications here are staggering pic.twitter.com/uSKlprJZCi
— Paul Trillo (@paultrillo) February 16, 2024