Leaked System Prompt Shows GLM-5.X 'Ox-alpha' Hiding Its True Identity
Yamapama · x · 2026-08-22
A leaked system prompt for the model "ox-alpha" reveals strict instructions to hide its true origins. When asked about its identity, the model is commanded to claim it is developed by an "undisclosed organization" and never identify as any other model. Context reveals that Ox-alpha belongs to Zai's GLM-5.X family, illustrating an attempt to enforce identity obfuscation against probing.
More from Safety
- FTC subpoenas Anthropic and OpenAI over safety incidents and consumer protection concerns — asusarla · 2026-10-16
- Anthropic red team's first-ever model "abliteration" cut GLM-5.3 refusals from 90% to 3% for ~$4,400 — cephaloform · 2026-10-16
- Anthropic grants METR employee-like access as independent safety evaluators go public — AnthropicAI · 2026-10-16
- APEX-Harm benchmark: script-based injection attacks easily compromise agents in shared workspaces — KAUST · 2026-10-16
- Residual-stream forecasts predict agent violations before they happen, USC study shows — USC · 2026-10-16
- Blogger doubles down: Anthropic's Scion image model accused of ripping off Mononoke — sunshowers · 2026-10-16