OpenAI appears to be stepping away from its high-cost venture into video generation. According to the Wall Street Journal, the company was spending approximately $1 million per day to maintain Sora. While this figure is substantial, users who were expected to celebrate the breakthrough have reportedly been leaving in large numbers. The base of enthusiasts has reportedly halved, from one million to five hundred thousand users, with no end to this decline in sight. Consequently, OpenAI has not only failed to retain its audience but seems to have abandoned a product that was draining its budget without clear prospects for return.
The challenges were not limited to waning interest. Sora faced more serious issues. Copyright concerns, while significant, were only part of the problem. Far more damaging to its reputation was the emergence of a deluge of low-quality, cliché-ridden content generated by the neural network. This could have transformed Sora from a breakthrough technology into a toxic asset, potentially undermining the OpenAI brand. The cancellation of even trial runs for new video models sent an unambiguous signal: internally, the company recognized that continuing down this path risked its reputation rather than creating value.
Amidst the ambitions of competitors such as Anthropic, OpenAI has opted for pragmatism. Instead of continuing to pour vast sums into complex video generation that has not yet justified its cost, the company has decided to reallocate its resources. The focus has shifted to code generation, enterprise solutions, and AI agents – areas with tangible revenue streams and clear business perspectives. According to unconfirmed reports, the Sora team will now focus on robotics, and the video product itself faces obscurity, with API access set to be discontinued in September. This represents more than just the closure of a project; it serves as a cautionary lesson: visionary thinking is important, but when it depletes budgets without tangible results, it is time to seek alternative paths.
Your audience is seeking genuine value, not just technological hype. CEOs should refrain from feeding investors with impressive demos. It is crucial to inquire about user retention metrics and actual return on investment (ROI), lest you risk repeating OpenAI's mistake by investing in hype rather than a viable product. The apparent struggles of Sora clearly demonstrate that even leading AI developers can err. Success in this field requires not only breakthrough technologies but also sound business planning capable of distinguishing a promising direction from an expensive distraction.