Sora's departure is a good thing for OpenAI and AI entrepreneurs. Big companies do big things! I just woke up in the morning and saw that a group of friends had transferred Sora to be offline. In about half a year, it became popular and accumulated a number of users. In my opinion, Sora is not a failed product, but it is not successful enough. Its growth rate has not kept pace with the increasingly fierce competition among major AI manufacturers. Sora's capabilities will be integrated into a super App that OpenAI is intensively developing. Disney, which has just provided IP licensing, also expressed understanding. For big companies like OpenAI, their previous strategy was to be big and comprehensive, pursuing all fronts. Two years ago, they probably did not pay attention to other AI manufacturers. It was also at that time that Anthropic put all their energy into the coding model. Maybe they saw something in advance, or maybe it was a helpless choice. Two years later in 2026, Anthropic basically had a huge impact on OpenAI. Google also came back with its own Gemini. China's open source and closed source large models are also booming. It seems that in just two years, the field of large models has flourished. At the same time, I originally thought that the coding model would only be used for coding, but the success of Manus has given great inspiration to all entrepreneurs and large companies, and has given VCs and entrepreneurs a clear and clear commercialization route. Now looking at all the large model sequences, the main focus must be coding capabilities, which support the current intelligent base of agents. Whether it is agent api, long context, harness engineering or long tasks, subagent, agent skills and other related capabilities and best engineering practices, a series of related capabilities and best engineering practices require solid coding capabilities. OpenAI is feeling a lot of pressure now, and is shrinking its strategy to deal with its competitors. It's not that Sora failed, it's because it has no extra energy to do side tasks. The opponents are too strong, so they can only continue to buy cards and build fiercely. This is in line with business logic, but the competition in the AI field is still too fierce. OpenAI in 2023 has just been familiar to the world, and it will be forced to give up its branch business in
- However, this so-called branch business is an area full of opportunities for many AI startups. From this point of view, it is a good thing. Big companies want to do big things.