Robotics evals should test hardware, not just software
techstatecraft.substack.com · 2,395 words · saved by 1 readers
Measuring the hardware overhang
A robot competing in the dexterous-hand competition at the 2026 World Humanoid Robot Games. Source: AGIBOT. I recently wrote about how frontier AI progress might generalize to robotics progress, leading to a rapid improvement in robot control capabilities. If frontier-AI progress substantially accelerated advances in robot control, true robot capabilities might be limited by robotic hardware – how precisely finger joints can move regardless of the digital inputs they receive, for example – rather than by software. It’d therefore be useful to know what capabilities robotic hardware could…
saved by
related reading
- How Claude Performs on Robotics Tasks \ Anthropicanthropic.com
- GitHub - adam-maj/robotics: A deep dive on the history of robotics and the future of humanoidsgithub.com
- How Can We Make Robotics More like Generative Modeling? | Eric Jangevjang.com
- Robot Dexterity Still Seems Hard - by Brian Potterconstruction-physics.com
- All Roads Lead to Robotics | Eric Jangevjang.com
- Demystifying evals for AI agents \ Anthropicanthropic.com
- The Final Offshoringfinaloffshoring.com
- Android Dreamsandroid-dreams.ai
- 14 Reasons Robotics is Hardsecondthoughts.ai
- Many Small Steps for Robots, One Giant Leap for Mankindnotboring.co
- Frontier of General-Purpose Robotics (2025)adampatni.com
- How do We Quantify Progress in Robotics? - by Chris Paxtonitcanthink.substack.com