Lessons from External Review of DeepMind's Scheming Inability Safety Case
Barrett, S., Zabala, F. J. C., Fillingham, S. P. , Siddique, U., Walpole, J., Bloomfield, R. 
ORCID: 0000-0002-2050-6151 & Papadatos, H. (2026).
Lessons from External Review of DeepMind's Scheming Inability Safety Case.
Abstract
Safety cases for frontier AI systems should provide a convincing argument, supported by evidence, that the risk of harm is within an acceptable bound. When developers author their own safety cases, confirmation bias and conflicted incentives can affect the quality of argument. External review can help to address this. In this paper, we apply the Assurance 2.0 framework to perform an external review of Google DeepMind’s public scheming inability safety case. We surface substantive new concerns that materially affect the scope of the safety case and its applicability for decision-making. Based on this experience, we provide concrete recommendations for how external review should be conducted and what information AI developers should provide to support it.
| Publication Type: | Article |
|---|---|
| Additional Information: | © 2026 This is an open-access pre-print distributed under the terms of the Creative Commons Attribution License (CC BY). |
| Subjects: | B Philosophy. Psychology. Religion > BJ Ethics H Social Sciences > HD Industries. Land use. Labor > HD61 Risk Management Q Science > QA Mathematics > QA75 Electronic computers. Computer science |
| Departments: | School of Science & Technology School of Science & Technology > Department of Computer Science |
| SWORD Depositor: |
Available under License Creative Commons Attribution.
Download (1MB) | Preview
Export
Downloads
Downloads per month over past year
Metadata
MetadataActions (login required)
Admin Login