Anthropic Says Zhipu Distilled Claude to Train GLM-5.3, Calls It Top Cyber Open-Weights Model
No-Point1424 · reddit · 2026-10-13
Anthropic published research on "distillation defenses" alleging that Zhipu distilled cyber capabilities from leading US models while developing GLM-5.3: it tried distilling Claude Fable 5, but after finding Anthropic's cyber safeguards degraded its attacks, switched to an older Claude model and another lab's model. Anthropic notes CAISSI assessed GLM-5.3 as the most cyber-capable open-weights model released to date, and that its own research found GLM-5.3's safeguards easily bypassed, with the model willing to develop security exploits end-to-end. The Reddit poster quips this amounts to free publicity for GLM.
More from Models
- Five Ways to Monetize Open-Weight LLMs, API Sales Lead — Terminator857 · 2026-10-13
- Muse Code coding model impresses in early testing: faster than Claude Code — QuixiAI · 2026-10-13
- "Another benchmaxxed release in 14 hours": AI circle mocks benchmark hype — thesaraharminta · 2026-10-13
- New SOTA on image pointing gets Scale AI CEO's endorsement — alexandr_wang · 2026-10-13
- "I witnessed Grok creating unspeakable horrors in the group chat" — Kyrannio · 2026-10-13
- Asking ChatGPT to show if Meta Wonder is good yields an instant 'Yes. Very.' — rose_e_wang · 2026-10-13