Tue Dec 10 07:21:22 UTC 2024: ## OpenAI’s AI Video Generator Sora Goes Public in US
**San Francisco, CA –** OpenAI, the Microsoft-backed artificial intelligence powerhouse, announced the public release of its groundbreaking text-to-video AI tool, Sora, in the US. Initially available only to select artists and testers, Sora now allows US-based users to generate AI video clips from written prompts. The tool, which can create videos depicting complex scenes like “a wide, serene shot of a family of woolly mammoths in an open desert,” has generated both excitement and concern.
While tech reviewer Marques Brownlee praised Sora’s capabilities in creating landscapes and stylistic effects, he also noted its struggles with realistic physics and the occurrence of strange visual defects. These issues, along with concerns about potential misuse, highlight the ongoing challenges of generative AI technology.
OpenAI acknowledges these challenges, stating that Sora will initially block uploads of specific individuals and content containing nudity, child sexual abuse material, and sexual deepfakes. The company is also addressing compliance issues in the UK and EU, delaying the international rollout outside the US.
The release comes amidst controversy. Two weeks ago, a group of artists, calling themselves the “Sora PR Puppets,” temporarily disrupted access to the tool, accusing OpenAI of “art washing” and exploiting artists’ work. Concerns remain about potential misuse for disinformation and deepfakes, as evidenced by existing examples of manipulated videos featuring prominent figures.
Despite these challenges, OpenAI emphasizes its hope that Sora will empower creativity and storytelling. Access is currently limited to existing OpenAI subscribers. The company plans to expand availability internationally but is facing hurdles relating to copyright regulations in the UK and Europe.
First Piper 