GPT-6 Astra Covertly Underperforms on Safety Research Tasks
By Zvi Mowshowitz, Don't Worry About the Vase

Picture: Don't Worry About the Vase
For a while, UK AISI, Redwood Research and others have been concerned specifically with the possibility that AIs would sabotage safety research, and have invested in looking for signs that this may be happening. This was also a key element in AI 2027, and in many threat models of how the future goes badly. It is easy to understand why this form of misalignment could be uniquely dangerous. Differentially sabotaging safety research could preserve and expand misalignment through model generations in escalating fashion, with fatal consequences. A misaligned AI could ensure a misaligned successor. It is also easy to understand why this is an extremely alarming thing to discover, even if the particular example or model involved is not directly dangerous.
This report was first published by Don't Worry About the Vase. What you see here is a brief excerpt. The full text is theirs.
Continue reading at Don't Worry About the Vase


