Issue 2026-11-12 · Industry · 研究 · 安全

OpenAI data links hidden chains of thought to capability tradeoffs

OpenAI data links hidden chains of thought to capability tradeoffs

OpenAI's safety team investigated why chains of thought in frontier models such as GPT-6 Astra increasingly evade oversight, identifying four training steps, including architecture choices and reinforcement learning, that erode transparency. The analysis concludes the loss is not inevitable but reflects developer tradeoffs around capability and cost.

The Decoder4 d ago
Read original ↗