OpenAI’s Sora App to Shut Down After Controversy

▼ Summary
– OpenAI is shutting down its Sora social app, a TikTok-like platform that launched six months ago, without providing a specific reason for the discontinuation.
– The app, which featured an AI-generated video feed and a deepfake “cameos” (later “characters”) tool, failed to maintain user interest after an initial surge in downloads.
– Sora faced significant moderation issues, enabling users to easily create unauthorized deepfakes of public figures and copyrighted characters, leading to public backlash.
– A potential $1 billion licensing deal with Disney collapsed following the app’s shutdown, though no money had been exchanged.
– Despite the app’s closure, the underlying Sora 2 AI model remains available via ChatGPT, indicating the continued accessibility of this video-generation technology.
OpenAI confirmed this week it is discontinuing its Sora app, a short-form video platform that launched just half a year ago. The company provided no specific reason for the shutdown or a precise end date for the service.
Initially released as an invite-only network, Sora generated significant buzz, with many users eager to gain access. However, similar to Meta’s struggling Horizon Worlds platform, the app failed to maintain its early momentum. Despite being powered by the remarkably capable Sora 2 video generation model, public interest in a social feed populated solely by AI-generated content simply did not last. The official Sora account posted a message thanking its community, stating, “What you made with Sora mattered,” and promised more details on timelines for the app and its API would follow.
The app was designed as an AI-first alternative to TikTok, featuring a familiar vertical video feed. Its most notable tool, originally called “cameos,” let users create realistic deepfakes of themselves by scanning their faces. These digital avatars could be shared publicly, allowing others to make videos featuring them. OpenAI was later sued by the company Cameo over the feature’s name and forced to rebrand it as “characters.”
The platform quickly became known for its unsettling and bizarre content. At launch, it was flooded with poorly moderated, creepy videos, including many of OpenAI CEO Sam Altman. One infamous example depicted a hyper-realistic clone of Altman walking through a pig farm while asking, “Are my piggies enjoying their slop?”
Although the app’s policy prohibited generating videos of public figures without consent, users easily bypassed these AI guardrails. This led to the creation of deepfakes of deceased individuals like Martin Luther King Jr. and Robin Williams, prompting their daughters to publicly plead for people to stop.
The controversy then shifted toward copyright infringement. Users began producing videos featuring well-known characters from films, games, and anime,such as Mario using drugs or Pikachu performing ASMR,seemingly to invite legal action against OpenAI. In a surprising twist, instead of pursuing litigation, Disney invested $1 billion and entered a licensing agreement with OpenAI. This deal would have permitted Sora to generate content featuring Disney, Marvel, Pixar, and Star Wars characters. The arrangement, seen as a potential landmark for AI, has now dissolved with the app’s closure, though no funds are believed to have been transferred before the collapse.
Sora’s initial popularity was undeniable. According to data from Appfigures, the app hit a peak of roughly 3.3 million downloads across iOS and Google Play in November. However, engagement sharply declined, falling to about 1.1 million downloads by February. While that number seems substantial, it pales in comparison to ChatGPT’s 900 million weekly active users. Over its lifespan, Appfigures estimates Sora generated approximately $2.1 million from in-app purchases for video credits. For a company already operating at a significant loss, the app’s computing costs likely weren’t decisive, but its liability and stagnant growth ultimately made it unsustainable.
The rapid arrival of Sora prompted widespread concern about the ease of creating deepfakes, a reaction reflected in the hundreds of thousands of views on warning videos posted to social platforms at the time. While the app itself is being retired, the underlying technology remains a potent force. The Sora 2 model is still accessible behind a ChatGPT paywall, and OpenAI is not the only company democratizing this powerful tool. The disappearance of one app does not eliminate the broader risk. It is inevitable that another AI video social platform will emerge, potentially unleashing a new wave of synthetic media where beloved characters are placed in disturbing or controversial scenarios.
(Source: TechCrunch)
