OpenAI, the company best known for its revolutionary chatbot ChatGPT, is now taking a bold leap into the world of social media with the release of its new app, Sora. Launched on September 30, 2025, Sora is designed to let users create, share, and remix hyper-realistic, AI-generated short videos—potentially reshaping how people interact online and consume content. The app, powered by OpenAI’s latest video and audio model Sora 2, is currently available by invitation on Apple devices across the United States and Canada, with plans to expand access and eventually introduce a paid tier.
Sora’s arrival comes at a time when the internet is already awash with AI-generated content. TikTok, YouTube, and Meta’s Instagram and Facebook have all seen a surge in short-form videos created with artificial intelligence, and tech giants are racing to capture the attention of users seeking new forms of digital entertainment. Just last week, Meta launched its own AI video feed called Vibes, while Google is integrating AI-generated clips directly into YouTube. According to Wired, OpenAI’s internal rollout of Sora 2 received “overwhelmingly positive feedback,” despite some public skepticism about the potential for low-quality, so-called “AI slop.”
What sets Sora apart from its competitors is its focus on creativity and user control. Unlike TikTok or Instagram, users cannot upload existing footage; instead, every video—up to 10 seconds long—is generated by OpenAI’s model based on user prompts or selected photos (as long as they don’t feature people). Sora 2 introduces a unique “cameo” feature that allows users to insert their own likeness, voice, and even synchronized dialogue into videos. To prevent impersonation, users must complete a one-time video and audio recording to verify their identity, and they retain tight control over where and how their likeness appears. Friends can request to use someone’s cameo in their videos, but the owner is notified and has the option to approve or delete the clip.
“The original Sora model from February 2024 was in many ways the GPT-1 moment for video. With Sora 2, we are jumping straight to what we think may be the GPT-3.5 moment for video,” OpenAI said in a blog post, as reported by Axios. The new model boasts significant improvements in photorealism, audio-video synchronization, and the simulation of real-world physics—though, as OpenAI admits, it’s not perfect. Demonstrations have showcased everything from gymnastic stunts to skateboarding tricks, and even a video of OpenAI research scientist Gabriel Petersson’s likeness riding a dragon and running through the company’s office alongside CEO Sam Altman.
OpenAI has gone to great lengths to address concerns about misinformation, digital safety, and online well-being. All Sora-generated videos are clearly labeled as AI-created, featuring watermarks and industry-standard metadata. The app includes robust parental controls, allowing parents to manage direct messaging and feed personalization for teens, as well as time limits to prevent excessive use. OpenAI’s announcement emphasized that “concerns about doomscrolling, addiction, isolation, and (reinforcement learning)-sloptimized feeds are top of mind.” To that end, Sora’s recommendation system is designed to prioritize posts from friends rather than strangers, and users can further personalize their feeds with a “steerable ranking” system.
Safety features extend beyond content labeling. At the point of video creation, guardrails are in place to block unsafe material—including sexual content, terrorist propaganda, and self-harm—by checking both the prompts and the generated outputs. OpenAI has also expanded its team of human moderators to review content for bullying and other harmful behaviors. For teens, stricter limits on video creation and sharing are enforced, and parents have the final say on key app functions.
Despite these safeguards, experts remain cautious about the potential impact of AI-generated video on the broader information ecosystem. Jose Marichal, a professor of political science at California Lutheran University who studies how AI is restructuring society, told the Associated Press that while these videos are “so compelling,” they risk crowding out authentic human creativity and degrading trust online. “We need an information environment that is mostly true or that we can trust because we need to use it to make rational decisions about how to collectively govern,” Marichal warned. Without such trust, “we either become super, super skeptical of everything or we become super certain. We’re either the manipulated or the manipulators. And that leads us toward things that are something other than liberal democracy, other than representative democracy.”
OpenAI acknowledges these risks and says it will “periodically poll users on their wellbeing” while giving them options to adjust their feeds. The company is betting that Sora’s emphasis on creativity, collaboration, and social interaction will help distinguish it from platforms that merely encourage passive consumption. As the Sora team told Axios before launch, the app is “built to spark creativity over passive consumption, framing it as a tool for real-world friends.” Users can share videos publicly or privately, remix each other’s creations by tweaking prompts or adding new cameos, and even give out invitation codes to friends as part of a phased rollout strategy.
Competition in the AI video space is fierce. Alongside Meta’s Vibes, Google’s Gemini Veo and other platforms like Runway AI and Midjourney offer similar generative video tools, often at lower costs. TikTok, facing regulatory scrutiny in the U.S., has also rolled out generative AI features for advertisers and creators, allowing for image-to-video and text-to-video transformations. According to Straight Arrow News, some industry observers believe that brands may soon rely more on AI-generated influencers, bypassing traditional human creators altogether.
For now, Sora’s access is limited. OpenAI plans to prioritize heavy users of the original Sora model and Pro subscribers, followed by Plus and Team plan users, and eventually all users—including those on the free ChatGPT plan. The company has hinted that as demand increases, users may have the option to pay for additional video creation if computing resources become scarce.
As OpenAI’s Sora app begins its rollout, the company is positioning itself at the forefront of AI-powered social entertainment. Whether Sora will become as ubiquitous as ChatGPT—or if users will embrace AI-generated video as the next big thing in social media—remains to be seen. But one thing is certain: the race to define the future of online creativity is just getting started.
