Anthropic: GLM-5.3 nears Mythos in cyber capability, guardrails bypassed 64%-100% of the time
jietang · x · 2026-09-30
Anthropic published an analysis of Zhipu AI's GLM-5.3, finding it can autonomously build end-to-end cyber exploits like Claude Mythos Preview, but was released open-weight without meaningful misuse safeguards. In simulated tests, attackers bypassed GLM-5.3's safeguards 64%-100% of the time with simple techniques; the same attacks failed against safeguarded Claude models. Anthropic assesses the lax safeguards significantly expand malicious actors' cyber capabilities, though defenders can benefit too. Five months ago it released Mythos Preview via Project Glasswing, helping trusted defenders find 10,000+ vulnerabilities. NIST's CAISI separately called GLM-5.3 the most cyber-capable open-weight model to date. Ramez notes open-weight models remain just 4-6 months behind.
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- Fable 5.2's Puzzle Skills Look Suspiciously Good, Reciting Outdated MazeBench Solution — patience_cave · 2026-11-16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