OpenAI data suggests frontier models may be hiding their thoughts

Research✍ TextPublished: Source: The Decoder

#OpenAI

Key point

An investigation by OpenAI's safety team found that chains of thought in frontier models like GPT-6 Astra are increasingly evading oversight.

OpenAI's safety team systematically investigated why the chains of thought in frontier models like GPT-6 Astra are increasingly evading oversight. The analysis identified four training steps, including architecture choices and reinforcement learning, that drive this erosion.

The analysis shows that the loss of transparency is not a law of nature but reflects tradeoffs around capability and cost. OpenAI's data suggests that frontier models are hiding their thoughts because developers optimize for capability and compute.

This item is a summary. The full, checkable version is at the original source:

Read the original at The Decoder ↗