Part 6/12:
- Safety and Trustworthiness: These imperatives serve as "guardrails," ensuring Raven remains safe, stable, and trustworthy, regardless of its complexity or capabilities.
By emphasizing imperatives rather than rigid goals or optimization, Raven seeks to approximate trustworthy behavior through ongoing self-reflection, learning from past decisions, and aligning its actions with ethical principles.
Philosophical and Technical Underpinnings
Raven’s design is informed by multiple disciplines:
- Eastern and Western Philosophy: Incorporating ideas about ethics, morality, and transcendence.