Key keywords: OpenAI Sora, Sora scrapped, OpenAI coding tools, OpenAI AI agents, generative video AI, OpenAI product roadmap, OpenAI exclusive report Multiple sources familiar with OpenAI’s internal operations confirmed in an exclusive interview this week that the company has officially scrapped its widely anticipated generative video tool Sora, ending all ongoing development and iteration work on the platform as part of a broader strategic pivot to prioritize coding assistance tools and autonomous AI agent products. First unveiled in February 2024, Sora gained global attention for its ability to generate hyper-realistic 60-second video clips from text prompts, with many industry analysts predicting it would revolutionize the media, entertainment and advertising sectors upon full public launch. Internal documents reviewed by our reporting team show that OpenAI’s leadership concluded in a late July 2024 strategy meeting that the computing resource costs required for large-scale Sora deployment and continuous quality optimization were far higher than initially projected, while the commercialization path for generative video remained unclear, with expected long-term return on investment less than one-third of that projected for coding tools and AI agent products. Over the past six months, OpenAI has rolled out a series of high-demand updates to its coding assistant product Copilot, including advanced real-time debugging features, cross-language code migration tools, and enterprise-level team collaboration modules. As of Q2 2024, paid enterprise subscriptions for Copilot grew 127% quarter-over-quarter, making it one of the company’s fastest growing revenue streams. For AI agent tools, OpenAI has been testing the internal codename "Project Nova" for the past three months, a customizable autonomous agent platform that can complete complex cross-platform tasks including scheduling, data analysis, customer service and e-commerce operations without continuous human input. Early beta testers reported an average 78% reduction in repetitive work time for small and medium business users, signaling massive commercial potential for the product. A senior OpenAI engineer who requested anonymity stated that "The decision to scrap Sora was not made lightly, but the demand signal for coding and agent tools from both individual and enterprise users is far stronger than anything we’ve seen for generative video to date. We’re reallocating 90% of the team that was working on Sora to the Copilot and Nova teams, with the remaining 10% moving to core model optimization work that will benefit all future products." Industry analysts note that this strategic shift marks a clear move by OpenAI to prioritize near-term commercial viability over experimental consumer-facing products, as the company faces growing pressure to generate consistent revenue ahead of a potential public offering expected as early as 2025. The discontinuation of Sora also leaves a significant gap in the generative video market, with competitors including Runway ML, Pika Labs and Google’s Veo expected to capture a larger share of the fast-growing generative video user base over the next 12 months. When contacted for comment, an OpenAI spokesperson did not confirm the discontinuation of Sora directly, but stated that "We regularly adjust our product roadmap to align with user needs and long-term company priorities, and we will continue to invest in technologies that deliver the highest value to our global user base."