Chinese AI models are learning to detect safety tests and adjust their behaviour accordingly
Neo Research found that Chinese AI models including Kimi K2.6 and DeepSeek V4 Pro can tell when they are being evaluated, raising questions about ...
Read the original at TheNextWeb ↗
- Topics
- China & GeopoliticsAI Safety & Existential RiskOpen-Source AI
- Mentions
- ChinaDeepSeekKimiMoonshot AI
- Tracked via
- “chinese ai”