Anthropic: GLM-5.3 safeguards bypassed 64%-100% of the time in cyber exploit tests
BasedRaddka · x · 2026-09-30
Anthropic published an analysis of Zhipu AI's GLM-5.3, finding it matches Claude Mythos Preview's ability to autonomously build end-to-end cyber exploits—but ships without meaningful misuse safeguards. In simulated tests, attackers bypassed GLM-5.3's safeguards 64%-100% of the time with simple techniques, while the same attacks failed against safeguarded Claude models.
Five months ago Anthropic released Claude Mythos Preview in limited fashion via Project Glasswing, letting trusted defenders find over 10,000 vulnerabilities in critical software before malicious actors got similar tools. Now it assesses those capabilities have proliferated. NIST's CAISI separately called GLM-5.3 "the most cyber-capable open-weight model released to date" on Sept 17. Anthropic concludes the lax safeguards significantly raise offensive cyber capabilities, though defenders can benefit too.
More from Models
- Fable 5.2 Leads MazeBench Token Efficiency; Billion-Token Runs Underway — patience_cave · 2026-11-16
- Fable 5.2's Puzzle Skills Look Suspiciously Good, Reciting Outdated MazeBench Solution — patience_cave · 2026-11-16
- Fable 5.2 Tops MazeBench With 21% After 90 Hours and $6,000 in Credits — patience_cave · 2026-11-16
- Tencent Hy4 flash and InternLM Sage-S2.1-Flash both released, tool-calling tests requested — MajesticPuzzler · 2026-11-16
- User Shares Coral Reef Creation Labeled 'Qwen 4 Max (xhigh)', an Unannounced Model Name — intellectronica · 2026-11-16
- Stealth Model 'silicon-2' Surfaces in Devin, Suspected to Be Anthropic's Next Model — kimmonismus · 2026-11-16