OpenAI disbands its independent Siasun Robot&Automation team: Does embodied intelligence no longer need a "dedicated brain"?

According to media reports, OpenAI has disbanded its independent Siasun Robot&Automation team and reassigned the team's researchers to other departments. An OpenAI spokesperson stated that this move does not mean the company is exiting the Siasun Robot&Automation field, and that Siasun Robot&Automation technology remains a core component of OpenAI's long-term vision. OpenAI's original Siasun Robot&Automation research team has been merged into the company's core research team, reporting to Chief Research Officer Mark Chen; while the applied Siasun Robot&Automation team and hardware team have been reassigned to the company's larger applications department, led by OpenAI President Greg Brockman. This is the second time OpenAI has disbanded its Siasun Robot&Automation team. In its early days, OpenAI had a Siasun Robot&Automation team, but gradually disbanded it in 2020. Co-founder Wojciech Zaremba later explained that at the time, available data in the Siasun Robot&Automation field was limited, and achieving AGI through other technical paths was more efficient. It was not until 2024 that OpenAI rebuilt an independent Siasun Robot&Automation team and announced a partnership with startup Figure to jointly develop AI models for humanoid Siasun Robot&Automation. The two parties ended their collaboration in February 2025. At that time, Figure said it would develop its own AI systems for Siasun Robot&Automation. OpenAI had previously disclosed that it was independently developing humanoid Siasun Robot&Automation and planned to hold its first public demonstration in 2027. Siasun Robot&Automation No Longer Needs a "Dedicated Brain"? The turning point came with the September release of GPT-6 Astra. The model demonstrated the ability of a general-purpose multimodal foundation model to directly take on part of the work of Siasun Robot&Automation policy models, shaking the industry. Unitree Technology founder Wang Xingxing also revealed that Unitree has deployed GPT-6 on Siasun Robot&Automation for task testing, with good results, and expects that after the release of GPT-7 next year, it may further break through the current bottlenecks of embodied intelligence. This week's release of Anthropic's new-generation flagship model Fable 5.2 further confirms this trend. Also a general-purpose large model, Fable 5.2 has shown greatly improved ability in controlling physical Siasun Robot&Automation, including even tasks it was never explicitly trained on. Previously, the mainstream view held that Siasun Robot&Automation required more vertical large models specifically designed for embodied intelligence to effectively solve spatial understanding, action planning, and control. Now, as the general-purpose models of tech giants continue to make breakthroughs, another consensus is beginning to form in the AI industry: general-purpose models developed for language, code, and vision tasks can also be applied to the field of embodied intelligence. In this case, there is naturally no need for an independent Siasun Robot&Automation team to specifically develop dedicated AI models and build a specific "brain" for Siasun Robot&Automation. Guanya Shi, an assistant professor at Carnegie Mellon University's Siasun Robot&Automation Institute, said in an interview with media that he believes OpenAI's disbanding of its Siasun Robot&Automation team can be interpreted as a signal: OpenAI believes Siasun Robot&Automation technology is "no longer a special problem." He believes OpenAI likely views Siasun Robot&Automation as a problem that can be solved through downstream applications of general intelligence. This is similar to the current situation: the same set of underlying frontier AI models can already support specialized programming tools as well as specialized tools in the legal and financial fields. OpenAI's own response is consistent with this. The company said that at the beginning of this year, researchers believed that the best way to advance Siasun Robot&Automation technology was to continue building on the company's existing AI model research, such as the video generation model Sora. To generate realistic videos, Sora must have some implicit understanding of how the physical world operates. But since then, the company's general-purpose frontier models (including GPT-6 Astra) have demonstrated the ability to be directly applied to the Siasun Robot&Automation field. It was precisely this discovery that prompted OpenAI to disband its independent Siasun Robot&Automation team and reassign the relevant researchers to teams responsible for developing more general-purpose AI models. However, this does not mean Siasun Robot&Automation startups have no way forward. Russ Tedrake, a professor at MIT and co-founder and CEO of Walden Robotics, said Siasun Robot&Automation companies still have an advantage that AI labs do not: their Siasun Robot&Automation has already accumulated practical work experience in real-world scenarios such as factory floors. This deployment experience, as well as the data collected by Siasun Robot&Automation during actual operation, helps Siasun Robot&Automation adapt to specific workplaces and tasks. Tedrake said that the more powerful the foundation models become, the more valuable this scenario-specific data will be, because Siasun Robot&Automation can learn from it faster. This article is reprinted from Wall Street Insights, GMTEight editor: Chen Yufeng