An Approach to Technical AGI Safety and Security (2025)¶
Reference
Citation: Google DeepMind. "An Approach to Technical AGI Safety and Security" (2025). arXiv:2504.01849. Type: report. Link: arxiv.org/abs/2504.01849.
What it is¶
A report on technical approaches to AGI safety and security. It sets out system-level controls that assume the model cannot be fully trusted.
Role in the record¶
- Grounds BP04: system-level controls (access control, monitoring, action restrictions) treating the model as untrusted.
Atom-level for/against detail and quotes are in the provenance data (assets/provenance.yml), keyed by practice atom.