Generative AI has been improving at a terrifying rate, with prompt-based images going from hilariously glitchy to often photorealistic in the space of mere months. Until now, AI generated video was lagging behind – but that all changed last week with the advent of Sora AI.
Created by OpenAI (of ChatGPT fame), Sora AI can create videos of up to 60 seconds "featuring highly detailed scenes, complex camera motion, and multiple characters with vibrant emotions". OpenAI shared various clips generated by Sora on X (below), and they're remarkably realistic.
Introducing Sora, our text-to-video model.Sora can create videos of up to 60 seconds featuring highly detailed scenes, complex camera motion, and multiple characters with vibrant emotions. https://t.co/7j2JN27M3WPrompt: “Beautiful, snowy… pic.twitter.com/ruTEWn87vfFebruary 15, 2024
The tech has already drawn awe and consternation online, but perhaps the most impressive thing here is just how much AI generated video has improved in one year. This time last year, the internet was collectively terrified by this nightmarish AI-generated clip of Will Smith eating spaghetti. As many have pointed out (below), that looks like child's play compared with these ultra-realistic new clips.
This is getting out of hand!- Will Smith pic.twitter.com/hHxqB07xC1February 19, 2024
🚨text to video progress in one year🤯2023: 2024: pic.twitter.com/gK57GmFAHvFebruary 15, 2024
Indeed, last year we were impressed by how much AI generated photos had come on in one year, and now it looks like its video's turn to enjoy the same accolade. But of course, as the most terrifying deepfakes have already shown, the potential for misinformation is pretty worrying when it comes to fake video – so the fact that content can now be
