It's probably because at every step, we always think that only a strong AI could solve the puzzle. So, when someone manages to make a not-strong-AI solve it, w

It's probably because at every step, we always think that only a strong AI could solve the puzzle. So, when someone manages to make a not-strong-AI solve it, we say they worked around the implicit rules and created a 'hack'.