Jumping over AI's Uncanny Valley / hacker news