OpenAI、SoraアプリとAPIを終了もChatGPT内でビデオモデル存続を示唆
OpenAIはSoraのアプリとAPIを終了し、ディズニーとの契約も破棄したが、同社はSoraのビデオ生成モデルがChatGPT内で存続することを示唆している。
キーポイント
SoraアプリとAPIの終了
OpenAIはSoraの専用アプリとAPIサービスを終了することを発表した。
ディズニーとの契約破棄
昨年12月に結ばれたディズニーとの10億ドル規模の契約が、ディズニー側から離脱されることになった。
ChatGPT内での存続
OpenAIは、Soraのビデオ生成モデル自体はChatGPTプラットフォーム内で存続し続けることを示唆している。
影響分析・編集コメントを表示
影響分析
この動きは、OpenAIが生成AI技術の提供形態を、独立した専門サービスから汎用プラットフォーム(ChatGPT)への統合へとシフトしている可能性を示唆する。また、主要企業との大規模契約の破棄は、生成AIビジネスの実用化と収益化における課題を浮き彫りにしている。
編集コメント
技術の進歩だけでなく、その提供形態とビジネスモデルの進化も注目すべきポイント。ChatGPTへの機能統合は、プラットフォーム戦略の一環として解釈できる。

OpenAIはSoraアプリとAPIを終了させています。ディズニーは、昨年12月に両社が締結したばかりの10億ドル規模の契約から撤退することになりました。
この記事「Sora's app and API are dead but OpenAI hints the video model lives on inside ChatGPT」は、The Decoderで最初に公開されました。
原文を表示
OpenAI is shutting down its Sora app and API. In response, Disney is pulling out of the partnership the two companies signed just last December.
OpenAI has announced that it's discontinuing its Sora AI video app. "We are saying goodbye to the Sora app. To everyone who created with Sora, shared it, and built a community around it: thank you," the Sora team wrote. The company says it will share more details soon, including timelines for the app and API shutdown, along with instructions for backing up any work users created.
It's still unclear what this means for OpenAI's broader video AI plans. The company initially posted a farewell to "Sora" entirely, then edited the wording to "Sora app." That distinction suggests the underlying Sora video model isn't completely dead.
The original post said goodbye to "Sora" entirely…
… then OpenAI changed it to "Sora App." | Image: Screenshot THE DECODER
Recently, there have been rumors that Sora will be integrated into ChatGPT as a video feature, as part of a planned "super app." According to the Hollywood Reporter, that's still the plan. The Information, however, reports that the idea has been shelved.
Disney walks away from a billion-dollar OpenAI deal
The fallout was immediate: Disney is pulling out of the deal it signed with OpenAI in December. Disney had agreed to invest one billion dollars in OpenAI and license some of its characters for use in Sora. The goal was to integrate the technology into Disney+.
"As the nascent AI field advances rapidly, we respect OpenAI’s decision to exit the video generation business and to shift its priorities elsewhere," a Disney spokesperson told the Hollywood Reporter, adding that Disney wants to continue working with AI platforms to "find new ways to meet fans where they are while responsibly embracing new technologies that respect IP and the rights of creators."
Sora's turbulent history ends as a footnote
The Sora app only launched in fall 2025 and immediately rattled Hollywood. The platform allowed free use of established intellectual property and well-known actors' likenesses, forcing OpenAI to backpedal within days. Studios and talent eventually got more control over their rights and likenesses on the platform.
OpenAI likely had several reasons for pulling the plug on Sora. The app completely tanked in the app charts after a viral launch. OpenAI CEO Sam Altman said at launch that the app would be discontinued if it didn't satisfy users. That lack of traction was probably the deciding factor—OpenAI wouldn't be killing it if it were growing fast.
But other factors likely played a role too. OpenAI has been shifting its strategy toward B2B lately, where rival Anthropic has been gaining ground. Video models in particular burn through massive compute resources that could be redirected toward research, development, and more profitable enterprise products.
OpenAI CFO Sarah Friar framed the decision as a resource problem. "We just are facing a lack of compute," Friar told CNBC. "We're having to make those really difficult decisions. Often we hold back models, we don't release features. And this was an example of having to prioritize." She added that the move doesn't mean OpenAI is done with creative tools for good: "It's not a 'never.' It's just a, 'We have to make hard choices.'"
Legal pressure added to the mix. OpenAI faced early pressure over copyright issues, the technical challenges of content moderation, and fierce competition from Chinese video models that were already outperforming Sora at lower prices. Chinese providers also largely sidestep Western copyright rules. Google, meanwhile, has made strong progress in video with its Veo models and leads in image generation thanks to Nano Banana. OpenAI would have had to invest heavily just to keep up.
What happens to Sora's world model research
The big remaining question is what happens to the research thesis behind Sora. When OpenAI first unveiled the model in spring 2024, it described Sora as a "world simulator" that could help train advanced AI—a key step toward artificial general intelligence.
Of course, OpenAI can keep pursuing this research without the pressure of turning the video model into a commercial product. But OpenAI's own track record shows a direct link between research depth and product success.
According to an internal memo seen by The Information, the Sora research team will now focus entirely on long-term world model research ("systems that deeply understand the world by learning to simulate arbitrary environments at high fidelity"), with the goal of "automating the physical economy."
AI News Without the Hype – Curated by Humans
Subscribe to THE DECODER for ad-free reading, a weekly AI newsletter, our exclusive "AI Radar" frontier report six times a year, full archive access, and access to our comment section.
Subscribe now
関連記事
今日のまとめ
AI日報で今日の重要ニュースをまとめ読み