OpenAI gave a group of talented filmmakers and professional artists access to its Sora artificial intelligence video generator. The move fueled a wave of creative exploration as they experimented with the tool's potential.
OpenAI, led by Sam Altman, unveiled Sora in February. This revolutionary AI video tool has rapidly become a game-changer in the field. Previously only available within OpenAI, Sora's powerful video generation capabilities are now unleashed by a select group of artists and filmmakers.
Their creations mark the first public showcase of content generated by OpenAI's groundbreaking video tool. OpenAI tasked visionary creative directors, musicians, directors, and production companies to unlock the AI-powered video tool's full potential.
The result? An impressive array of realistic and surrealist visions pushes the boundaries of AI-generated video. The source material used to train Sora remains shrouded in secrecy. Concerns have been raised that YouTube content might have been included without proper consent from creators.
Despite lingering concerns about training data, a segment of the creative community is willing to embrace Sora's potential. Let's explore how creatives are using Sora. Filmmaker Paul Trillo sees Sora's true strength in realising new and impossible ideas.
Sora: A new era of AI video generation
"Sora is at its most powerful when you're not replicating the old but bringing to life new and impossible ideas we would have otherwise never had the opportunity to see," Trillo said in a blog post about Sora's first impressions on OpenAI's official website.
Trillo's creations delve into the obscure, taking viewers on whirlwind journeys through unconventional locations. Blending fantastical robots and dancing refuse with surreal scenes, his videos push the boundaries of visual storytelling.
Trillo's other Sora creation centres on an epic journey of the Voyager spacecraft's Golden Record. It took 11 different generations cut together to "explore what the odyssey of this record might look like."
Toronto's multimedia studio, Shy Kids, stole the show with a stunning short film, "Air Head." This short film uses a series of thematically linked Sora clips to tell the story of a man with a balloon for a head, narrated by a compelling voiceover.
OpenAl just released one of the first short films, “Air head”, created by the artists Shy Kids using Sora Al pic.twitter.com/8WaVliT3ZR
— Historic Vids (@historyinmemes) March 27, 2024
The project also highlighted areas for improvement in character consistency within Sora. OpenAI is expected to address these shortcomings before the AI tool's official launch.
How are creatives, artists using AI video?
AI video and generative AI video have existed for over a year. However, OpenAI's Sora appears to be a game-changer. Unlike its counterparts, Sora can craft longer, more immersive narratives from a single text prompt.
It's up: https://t.co/WaBYlrVzI0
— Paul Trillo (@paultrillo) March 26, 2024
The AI company attributes Sora's capabilities to a major processing power and training time leap. This significant leap forward has captured the attention of artists, pushing the boundaries of what AI video creation tools can achieve.
Nik Kleverov: Thrilled to unveil my journey with Sora - I can already see how it’s going to transform the way I approach agency work as well as personal projects. For the latter- it’s allowing me to iterate and explore original concepts that have been#SORA #OpenAi #openaisora pic.twitter.com/6rxD8dUxPS
— AI Community (@AIIZ_KR) March 27, 2024
OpenAI's first impressions series showcased the diverse benefits of Sora. Emmy-nominated agency Native Foreign's Nik Lkeverov visualised concepts for fictional soda brands. Musician August Kamp sees Sora as a tool to rapidly iterate on cinematic visuals, while creative director Josephine Miller of Oraar Studio praises its ability to breathe life into long-dormant ideas.
"My thoughts drifted towards exploring the realm of photogrammetry and its potential applications to sculpture," Alexander Reben, current OpenAI artist in residence, said. "The prospect of transforming video into 3D models intrigued me, as it hinted at propelling the AI system beyond its initial scope."
We did the impossible task and improved @DonAllenIII already masterpiece video - we dubbed (of course saving the original voice) it in French, Spanish, Chinese, and Hindi. What do you think? @OpenAI #Sora #SoraAI pic.twitter.com/bdbUyEm9pS
— rask.ai (@rask_ai) March 26, 2024
Altman recently emphasised that AI is not a sentient being but a powerful tool with specialised capabilities. OpenAI is also currently focusing on unlocking the secrets to a longer life.
