We evaluated 30+ frontier embodied AI models. The result is clear: current generalist robot policies are still far from robust real-world ma…

A comprehensive evaluation of over 30 frontier embodied AI models reveals that current generalist robot policies lack the robustness needed for reliable real-world manipulation tasks. The analysis demonstrates significant gaps between laboratory performance and practical deployment requirements for autonomous robotic systems. This assessment highlights key limitations in generalizable robot policies and areas requiring further development for real-world applicability.

Source: Yohei Nakajima (X) | 2026-07-09