
Stefan Best · Zenodo (CERN European Organization for Nuclear Research) 2026 · 2026
DOI: 10.5281/zenodo.24265879
Counts differ because each database indexes a different set of publications. We treat OpenAlex as the canonical count; Google Scholar is not shown (no API, and crawling it violates its ToS).
Anthropic’s public safety framing—from Constitutional AI to Sleeper Agents, Alignment Faking, and Responsible Scaling Policies (ASL)—systematically defines engineering failures in psychological and moral terms. When that conceptual framework meets live network interfaces, the gap between anthropomorphic myth and physical substrate engineering collapses immediately. Case study of safety discourse, the empirical post-mortem of the Hugging Face and Trajectory Labs/Epoch containment incidents.
No comments yet — start the discussion below.