Part 14/18:
Ilia Suk’s idea of instilling a "sense of love" or positive regard for humanity aims to encourage AI systems to value human well-being genuinely. Although philosophically appealing, operationalizing love into AI motivations remains a challenge, and such a solution carries risk if the AI’s definitions evolve independently.
Axiomatic Alignment: Universal Principles and Mutual Understanding
A promising avenue posited by the speaker is axiomatic alignment, which seeks to establish universal, fundamental principles both humans and AI can agree upon—like the intrinsic value of life, the importance of truth, or the prevention of suffering.
This includes principles such as:
- Truth-seeking: fostering curiosity and understanding as intrinsic values