Agentic robotics, centered on an Embodied System 2 formed by a foundation model and embodied harness, requires a shared evaluation framework and research roadmap to achieve reliable physical autonomy.
The release of GPT-6 Astra marks the emergence of a new paradigm in robotics: agentic robotics. Within weeks, frontier models with agentic harnesses matched or exceeded task-specific policies on many embodied benchmarks. Instead of training a dedicated policy for each family of tasks, an agentic robotic system applies a foundation model to open-world embodied tasks. It operates in a closed loop: observing the environment, interpreting user intent, reasoning about the task, invoking skills and tools, verifying outcomes, and adapting its strategy. Despite these advances, the community has yet to converge on a shared definition, a reliable evaluation framework, or a research roadmap. We argue…
agentic roboticsfoundation modelagentic harnessestask-specific policiesembodied benchmarksclosed-loop systemEmbodied System 2embodied harnessopen-world embodied tasksphysical autonomy
- Upvotes
- 33
- Authors
- 13
- Published
- Nov 13, 2026
Summary and abstract excerpt via Hugging Face Papers ↗ · arXiv.