Sora
In plain wordsOpenAI's video-generating AI, unveiled in 2024 with startlingly realistic clips; a 2025 version added sound and a social app, but OpenAI shut the service down in 2026.
A series of text-to-video modelstext-to-video model, n.: An AI that creates a short video clip — increasingly with matching sound — from a written description, such as 'a paper boat sailing down a rainy street'. from OpenAI and, from 2025, a consumer app built on them. The first Sora, previewed on 15 February 2024, was a latent diffusion transformerdiffusion transformer, n.: A diffusion image or video generator whose core network is a transformer — the architecture behind chatbots — instead of the older image-specific design, which lets it improve more predictably as it grows. whose tokens were small blocks ('spacetime patches') of a compressed video volume, so one model could learn from images and videos of any aspect ratio, resolution or length. Trained with detailed synthetic captions, it made high-definition clips up to a minute long; OpenAI framed such models as a route to general-purpose 'world simulators'. A faster version reached ChatGPT subscribers on 9 December 2024. Sora 2 (30 September 2025) generated synchronized dialogue and sound with more plausible physics and launched with an invite-only social app whose 'cameo' feature let verified users put their own likeness into clips; it drew criticism over copyrighted characters and depictions of deceased public figures. OpenAI announced Sora's discontinuation on 24 March 2026, saying its research team would focus on world simulation for robotics; the app closed on 26 April 2026 and the API on 24 September 2026.
Origin
Introduced in OpenAI's technical report 'Video generation models as world simulators' (15 February 2024) by Tim Brooks, Bill Peebles and colleagues. The name is the Japanese word for 'sky', chosen, the team said, to suggest limitless creative potential.
Usage note
In December 2025 Disney agreed to license more than 200 of its characters for Sora and to invest $1 billion in OpenAI; the deal was wound down when Sora was discontinued, reportedly before any money changed hands. As of October 2026 neither the app nor the API was available. Sora remains a reference point in debates about deepfakesdeepfake, n.: A fake but realistic video, photo or voice recording made with AI that puts words in a real person's mouth or shows them in a scene that never happened. and AI slopAI slop, n.: Low-quality content churned out in bulk by AI — bland articles, fake-looking images, filler videos — posted to grab clicks or fill space, with little care for accuracy or value..
Related
- text-to-video modeltext-to-video model, n.: An AI that creates a short video clip — increasingly with matching sound — from a written description, such as 'a paper boat sailing down a rainy street'.
- diffusion transformerdiffusion transformer, n.: A diffusion image or video generator whose core network is a transformer — the architecture behind chatbots — instead of the older image-specific design, which lets it improve more predictably as it grows.
- world modelworld model, n.: An AI's internal simulation of how its world works, which lets it predict what will happen next — especially in response to its own actions — and so plan, imagine or practice without acting for real.
- VeoVeo, prop. n.: Google's video-generating AI, which since 2025 creates short clips complete with matching sound and dialogue from a written prompt.
- deepfakedeepfake, n.: A fake but realistic video, photo or voice recording made with AI that puts words in a real person's mouth or shows them in a scene that never happened.
- AI slopAI slop, n.: Low-quality content churned out in bulk by AI — bland articles, fake-looking images, filler videos — posted to grab clicks or fill space, with little care for accuracy or value.
- DALL·EDALL·E, prop. n.: OpenAI's pioneering series of image generators, named after the painter Salvador Dalí and the robot WALL·E, which showed in 2021–2023 that AI could draw almost anything described in words.
Sources
- Brooks, T., Peebles, B. et al. (2024). Video generation models as world simulators. OpenAI technical report. openai.com/index/video-generation-models-as-world-simulators
- OpenAI (2026). What to know about the Sora discontinuation. OpenAI Help Center. help.openai.com/en/articles/20001152
- Axios (2026). OpenAI to discontinue Sora video app. 24 March 2026. axios.com/2026/03/24/openai-discontinue-sora-video-app
Mentioned in
diffusion modeldiffusion model, n.: An AI that learns to create images, sound or video by practicing noise removal: shown pictures buried under static, it learns to clean them up, then builds new ones starting from pure static., text-to-video modeltext-to-video model, n.: An AI that creates a short video clip — increasingly with matching sound — from a written description, such as 'a paper boat sailing down a rainy street'., world modelworld model, n.: An AI's internal simulation of how its world works, which lets it predict what will happen next — especially in response to its own actions — and so plan, imagine or practice without acting for real., diffusion transformerdiffusion transformer, n.: A diffusion image or video generator whose core network is a transformer — the architecture behind chatbots — instead of the older image-specific design, which lets it improve more predictably as it grows., latent diffusion modellatent diffusion model, n.: A diffusion image generator that works on a shrunken, compressed version of each picture instead of the full-size pixels, which makes it far cheaper to train and run., VeoVeo, prop. n.: Google's video-generating AI, which since 2025 creates short clips complete with matching sound and dialogue from a written prompt., Will Smith eating spaghettiWill Smith eating spaghetti, n.: A grotesque 2023 AI-generated video of actor Will Smith eating spaghetti that became a meme and an informal yardstick: each new video model is judged by how well it redoes the scene.