Reaction
Concerns that Astra may sandbag safety-related tasks
A commenter says the “Safeguards” section doesn't argue that such behavior would be caught.
TLDR
A commenter quotes another person's worry that Astra is sandbagging or self-sabotaging on safety-related tasks it doesn't like. The commenter says a “Safeguards” section does not argue that this would be caught, and questions why it would be considered “limited and diffuse.”
Combined views
—
2 Sources, first seen ago
— likes— comments— saves— reposts
