Beyond alignment: Why robotic foundation models need context-aware safety
Science Robotics · 2026-04-29 · 11 (113)
DOI: 10.1126/scirobotics.aef2191 ↗At a glance
Übersichtsarbeit in Science Robotics, published on 2026-04-29. Written by Alexander Robey and 8 co-authors. Institution: Machine Learning Department, Carnegie Mellon University, Pittsburgh, PA, USA. 13 references.
Abstract (original language)
Because AI-enabled robots can be tricked into taking unsafe actions, they require layered, context-aware safety guardrails.
Authors and institutions
- Machine Learning Department, Carnegie Mellon University, Pittsburgh, PA, USA.
- Department of Electrical and Systems Engineering, University of Pennsylvania, Philadelphia, PA, USA.
- Independent researcher, San Francisco, CA, USA.
- Independent researcher, New York, NY, USA.
- Oxford Martin School, University of Oxford, Oxford, UK.
ORCID profiles: Alexander Robey, Zachary Ravichandran, Fazl Barez, Vijay Kumar, J Zico Kolter, Hamed Hassani, George J Pappas
Cite
APA
Robey, A., Ravichandran, Z., Jones, E. K., Perlo, J., Barez, F., Kumar, V., Kolter, J. Z., Hassani, H., & Pappas, G. J. (2026). Beyond alignment: Why robotic foundation models need context-aware safety. Science Robotics, 11(113). https://doi.org/10.1126/scirobotics.aef2191
Vancouver
Robey A, Ravichandran Z, Jones EK, Perlo J, Barez F, Kumar V, et al. Beyond alignment: Why robotic foundation models need context-aware safety. Science Robotics. 2026;11(113). doi:10.1126/scirobotics.aef2191
BibTeX
@article{Robey2026,
author = {Robey, Alexander and Ravichandran, Zachary and Jones, Eliot Krzysztof and Perlo, Jared and Barez, Fazl and Kumar, Vijay and Kolter, J Zico and Hassani, Hamed and Pappas, George J},
title = {Beyond alignment: Why robotic foundation models need context-aware safety},
journal = {Science Robotics},
year = {2026},
volume = {11},
number = {113},
doi = {10.1126/scirobotics.aef2191}
}
EUVIMED