Andrew Kang, a prominent figure and co-founder of Mechanism Capital, recently asserted on social media that the "AlphaGo moment" for Physical AI in robotics has already been realized, signaling an imminent proliferation of intelligent robots. Kang's tweet, shared on July 29, 2025, predicts that these advanced robots will begin widespread deployment in production environments across various industries as early as 2026.
"Everyone talks about the ChatGPT moment for robotics, but the AlphaGo moment has already been realized for PhysicalAI and this is all that's needed for the initial proliferation of intelligent robots," Kang stated.
Physical AI refers to the integration of artificial intelligence with physical systems like robots, smart machines, and autonomous vehicles, enabling them to perceive, understand, and interact with the real world. Unlike traditional AI that operates solely in digital realms, Physical AI provides a "body" to the AI, allowing it to perform complex actions and adapt to dynamic environments. This fusion aims to revolutionize capabilities from industrial process optimization to healthcare.
The "AlphaGo moment" in AI signifies a pivotal breakthrough where an artificial intelligence system achieves superhuman performance in a complex domain, much like DeepMind's AlphaGo surpassed human champions in the game of Go. Kang suggests that Physical AI has reached a similar inflection point, where fine-tuned models are demonstrating superior performance and accuracy over human capabilities, even over extended periods.
"We are already at the point where some researchers/companies are developing fine-tuned models that achieve better performance and accuracy than a human over long time horizon," Kang elaborated.
Recent advancements in robotics support this claim, with examples such as surgical robots achieving superhuman precision and speed in tasks like suturing and needle insertion. Fine-tuning techniques, which adapt pre-trained AI models with smaller, specific datasets, are crucial for optimizing robot performance for particular tasks. While cross-task generalization remains a long-term goal, the ability to train robots to excel in even one task quickly creates significant demand.
The projected rollout of these robots in 2026 aligns with a surge in investment within the robotics sector. In the first half of 2025 alone, robotics startups secured over $6 billion in venture capital funding, driven by advancements in AI-integrated humanoids. Companies like Apptronik have recently closed significant funding rounds to push the commercialization of humanoid robots for applications in areas like warehousing and logistics, underscoring the industry's readiness for wider adoption.