The Reasons Behind Sora’s Shutdown by OpenAI

On Tuesday morning, everything seemed routine at OpenAI. By day’s end, however, the company had made some significant announcements: it would be discontinuing its video-generation app, Sora, halting plans for video generation within ChatGPT, winding down a $1 billion deal with Disney, reshuffling a high-level executive position, and securing an additional $10 billion from investors, bringing its latest funding round total to over $120 billion.

OpenAI is now in a race to achieve profitability or at least minimize losses. Since its inception, Sora has consumed a considerable amount of computational resources without delivering the financial returns to justify its existence. Sources indicate that it has been falling behind rival video-generation models. Despite its brief existence, Sora has contributed to a growing skepticism about the authenticity of digital content.

As OpenAI navigates investor scrutiny and fierce competition from other AI firms, executives seem to agree that a strategic pivot is necessary. “We cannot miss this moment because we are distracted by side quests,” said Fidji Simo, OpenAI’s CEO of AGI deployment, who recently transitioned from her role as CEO of applications. “We really have to nail productivity in general and particularly productivity on the business front.” This shift means stepping back from projects like Sora and reportedly deprioritizing the exploration of “adult mode” features for ChatGPT.

Sora was already facing challenges in the competitive AI video generation landscape, according to Trevor Harries-Jones, a board member at the Render Network Foundation, which allows creators to explore and compare AI-generated video. “The state of innovation and the plethora of choice means there’s just little to no moat and it’s very simple to switch between,” Harries-Jones noted. “If your model is not the top at any one thing, it’s very hard to get mass usership.”

Harries-Jones speculated that OpenAI’s decision stemmed from the rapid pace of innovation in the video-generation sector and intense competition from companies like Google and Kling. “For us, there wasn’t anything that they were winning, in terms of one of the use cases,” he said, indicating that Sora lacked a clear advantage. “It had fallen off from a very strong start, eclipsed by other players in the space.” He also pointed out that while the announcement and marketing videos seemed groundbreaking, there was a noticeable gap between the hype and the actual product launch. “As with all these, the devil is in the details on the cost, the timeframe that can be generated, and more,” he added.

Sora’s user engagement issues may be reflected in its download statistics. Seema Shah, VP of insights at a market intelligence firm, mentioned that Sora was “one of the fastest-growing apps when it first launched” but experienced a significant decline after a couple of months, likely due to competition. Sora started strong in October with around 4.8 million downloads globally, peaking at 6.1 million in November, but then saw a sharp decline in December (3.2 million) and subsequent months: 2.1 million in January, 1.4 million in February, and just 1.1 million so far in March.

“What’s most notable about it is dropping off while they’re expanding into new markets — that should be driving growth,” Shah observed, adding, “You should’ve seen an uptick in that. Even if nobody else in the US downloaded it again, there should be some growth, presumably.”

The video generation process consumed a lot of costly computational power at a time when OpenAI was in dire need of it. During the company’s annual DevDay event in October, executives emphasized the necessity of generating more revenue and expressed concerns that compute limitations could hinder OpenAI’s scaling efforts. “Obviously someday, we have to be very profitable,” OpenAI CEO Sam Altman stated, noting that the company was in a phase of “investing aggressively.”

On the same day, company president Greg Brockman remarked, “Asking, ‘How much compute do you want?’ is a little bit like asking, ‘How much of the workforce do you want?’ The answer is you can always get more out of more.” OpenAI also hinted at introducing ads within ChatGPT Pulse, which eventually materialized as ads throughout ChatGPT last month.

The revenue push has been further complicated by the loss of the Disney deal. Last year, Disney had promised a $1 billion equity investment in OpenAI, with the option to purchase additional equity, and stipulated that Disney would become a “major customer” of OpenAI, utilizing its products for new offerings on Disney+ and making ChatGPT available to employees. The deal also allowed Sora to feature AI-generated videos of numerous Disney, Pixar, Star Wars, and Marvel characters, with some available for streaming on Disney+. However, less than an hour after OpenAI and Disney began collaborating on a Sora-related project, Disney was caught off guard by the decision to discontinue the app.

The swift termination of the OpenAI-Disney partnership — just over three months into a three-year licensing agreement — surprised many. While some online discussions suggested it indicated a broader failure of AI in entertainment, an industry expert noted that Disney remains open to licensing agreements with other video-generation AI companies, potentially leading to partnerships with firms like Google, Runway, Luma, Moonvalley, Kling, or Seedance.

“We see a lot of momentum in licensing,” the expert said, adding that for major companies like Disney, such agreements are often strategic moves to foster fan engagement in new ways. “The Disney-OpenAI deal was one sign of that. I think it’s great that in the exit announcement, Disney made it clear that they are wide open for business to continue character licensing with other parties.”

The news of Sora’s shutdown came just a day after OpenAI published a blog post on using Sora safely, stating that the company was committed to “strengthening Sora’s guardrails” and implementing stricter protections for videos featuring children and teens. Earlier in the month, OpenAI had also planned to integrate Sora’s video-generation capabilities into ChatGPT before scrapping that strategy.

“We’ve decided to discontinue Sora in the consumer app and API,” OpenAI spokesperson Kayla Wood stated, indicating that more information on the timeline for winding down the app and API would be shared soon. “As we focus and compute demand grows, the Sora research team continues to focus on world simulation research to advance robotics that will help people solve real-world, physical tasks.”

OpenAI also mentioned the need to concentrate its existing compute resources on its AI agent goals, noting that Sora was a step toward enhancing world simulation research, particularly in robotics. Interestingly, the original announcement regarding Sora’s discontinuation was edited, possibly to set the stage for OpenAI’s ongoing work in world simulation AI, which may utilize the model behind Sora for further advancements.

As OpenAI shifts its focus from a flurry of revenue-generating initiatives — including ventures into social media, new browsers, subscription tiers, advertising strategies, and government contracts — and expands its resource output into coding and enterprise tools, it will face even stiffer competition from Anthropic. This rival has established a reputation for a clear enterprise focus and is reportedly leading the market in coding tools.

Sam Gregory, executive director at a nonprofit dedicated to combating deceptive AI and deepfakes, expressed no regret over the loss of a tool for generating “AI slop.” However, he is frustrated that “in the absence of real money, real investment, real will,” changes seem to occur only for business reasons rather than addressing genuine harm.

“Sora launched with a splash in terms of this ability to create hyper-realistic content… what angers me is what they normalized in the period of six months, a set of things that I think have really long-lasting implications,” Gregory stated. “They normalized a world in which in mere seconds, people are really uncertain about what they are seeing in their timelines, both when it matters and when it doesn’t… whether it’s casual comedy or conflict footage from Iran.” He added, “The consequences will reverberate even if the app is gone.”

Altman’s earlier assertion that the company was “investing aggressively” now appears to have shifted to a focus on appeasing investors, especially as OpenAI reportedly plans to go public as soon as this year. With billions of dollars continuing to flow into the AI sector, investors are increasingly scrutinizing their returns and assessing which segments of the industry may be overvalued.