Moonshot strips cyber capabilities from open-weight Kimi, keeps full version internal
robleclerc · x · 2026-10-24
Moonshot is taking an unusual approach to cyber safety for its new Kimi 3.1 Pro: instead of adding guardrails in post-training, it removed cyber-related data during mid- and post-training, aiming to prevent people from fine-tuning the safety away in the open-weight release. A more capable internal version with full cyber capabilities is retained. Commenters note this means China's true AI frontier is no longer public, mirroring the US frontier-lab model.
More from Models
- User reports dot's web fetch stays restricted after removing a custom connector — MotherIsland8888 · 2026-10-25
- $0.48 and 43 minutes: DeepSeek V4.1 builds an interactive map, igniting cost vs capability debate — teortaxesTex · 2026-10-25
- Do ISTA's 2-bit GSQ-RCO DeepSeek quants really match the base model? Community asks for KLD — Queasy_Back2588 · 2026-10-25
- Open-source DNA model Carbon-X (14.8B) nears frontier scores on genomics at 1/1800th the cost — aigclink · 2026-10-25
- MemT 1.0: 293B memory-native MoE puts persistent memory inside the forward pass — AKAMA199 · 2026-10-25
- GPT-X3-150M tops Open SLM Leaderboard, dethrones SmolLM2-135M with GDN2/GQA hybrid — datdanboi25 · 2026-10-25