$OpenAI(OpenAI.NA) disbanded its robotics team, not because it's abandoning robot R&D, but because the recently rele…
$OpenAI(OpenAI.NA) disbanded its robotics team, not because it's abandoning robot R&D, but because the recently released GPT-6 Astra outperformed vertical-domain large models in simulated robot tests. Rather than diluting GPU resources and R&D efforts on robots, it makes more sense to focus on developing general-purpose models. Back in May, $DeepSeek(DPSK.NA)'s Liang Wenfeng stated that the technical path to AGI is very clear: Language Model -> CoT -> Agent -> RSI. Once the technological singularity is reached, capabilities like multimodality, 3D, and world models will naturally emerge. This looks incredibly prescient now. Not only does Astra demonstrate stronger capabilities than vertical robot models, but Fable 5.2 can also construct SVG vector illustrations and interactive HTML 3D Sydney scenes via code. It should surpass existing multimodal technologies, including ByteDance's $ByteDance(BYTED.NA) Seedance, sooner rather than later.
The copyright of this article belongs to the original author/organization.
The views expressed herein are solely those of the author and do not reflect the stance of the platform. The content is intended for investment reference purposes only and shall not be considered as investment advice. Please contact us if you have any questions or suggestions regarding the content services provided by the platform.
