Computer Science > Software Engineering
Title:Safe AI Evaluation: A Systems Perspective
Abstract:AI evaluation can involve an active, adaptive optimisation process in which AI agents re-plan, retry, use tools or delegate work to achieve a goal. Unanticipated capabilities or unsafe behaviour may emerge, and existing controls may be inadequate to prevent harm. Selected safeguards may also be deliberately relaxed to observe behaviour they would otherwise suppress, creating additional risks. This article examines the distinctive safety considerations of evaluation and their implications for control placement, observation and intervention, and presents four architectural patterns to support safe AI evaluation, including diverse supervisors, an independent verifier, silent monitoring and coordinated shutdown.
| Subjects: | Software Engineering (cs.SE) |
|---|---|
| Cite as: | arXiv:2611.24987 [cs.SE] |
| (or arXiv:2611.24987v1 [cs.SE] for this version) | |
| https://doi.org/10.48550/arXiv.2611.24987 arXiv-issued DOI via DataCite (pending registration) |
Submission history
Access Paper:
Additional Features
- Audio Summary
Current browse context:
References & Citations
- Google Scholar
- Semantic Scholar
BibTeX formatted citation
Bookmark
Bibliographic and Citation Tools
Code, Data and Media Associated with this Article
Demos
Recommenders and Search Tools
arXivLabs: experimental projects with community collaborators
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.