Ethics of AI ·
Who is harmed and who is helped when systems learn to act
un
seel
.com · alignment · trolley · welfare · accountability
Alignment
—
Risk
—
State
—
Raw code · uninterpreted goal
Human-value filter
Aligned ethics node
Risk spark · misalignment
▶ Play
←
→
🔇 Unmute
Reset
Un
seel
.com · Ethics of AI