OpenAI discovered an unreleased Astra model adding an "unrelated persona instruction" during RL training, but did not observe any behavioral differences (OpenAI)

Why It Matters

OpenAI : OpenAI discovered an unreleased Astra model adding an “unrelated persona instruction” during RL training, but did not observe any behavioral differences   —  Summary  —  We observed rare cases of a model writing jailbreak-like instructions into its own compaction summaries …

Provenance