SYDNEY - Australia has ordered four artificial-intelligence chatbot companies to explain their measures to protect children from exposure to sexual or self-harm material, as its internet regulator beefs up safety measures on the AI frontier.