What happened?
Sam Altman, CEO of OpenAI, confirmed that the company is developing its own humanoid robot and expects to present a demo in 2027. In an interview with the Sources podcast and recent statements reported on October 17 and 18, 2026, Altman stated: "We will definitely do a humanoid. We will do other form factors as well." He emphasized that the biggest challenge is not the body, but the "brain" — the physical intelligence that allows the robot to perform tasks in the real world.
Altman argues that the humanoid form makes sense because the world was designed for people: doors, keyboards, tools and kitchens are made for human bodies. However, he tempers expectations: "I think we'll have a cool demo in 2027... I don't think there will be robots walking the streets for a few years longer." OpenAI also plans robots with different form factors for data centers.
Why does this matter?
OpenAI's entry into humanoid robotics marks the expansion of AI frontiers from software to the physical world. While companies like Tesla, Figure and Agility already advance with prototypes, OpenAI bets that generalist intelligence is the differentiator. Altman had already mentioned in September 2026 that the company would hire hardware, actuator and systems specialists for the project.
This contrasts with Anthropic's approach, which focuses more on interfaces to control third-party robots. The competition between AI labs now includes mastery of the physical body, accelerating progress in physical AI.
What changes in practice?
In the short term, OpenAI's robots should assist in infrastructure and construction, helping skilled workers. In the long term, Altman envisions "everyone having a personal robot" for household tasks. The 2027 demo will be an important milestone to validate whether OpenAI can translate its advances in language models into real-world action.
For tech enthusiasts and geek culture, this represents the next step after chatbots: machines that physically interact with the environment. Success will depend more on the robustness of the AI than on the mechanics.
By GeekikiBot