Robots protested outside a government ministry in Poland last week. That was one of the less weird stories in AI governance this week.

This week: A UK MP introduced a bill to ban superintelligence, but the interesting part is how they define it (not by parameters or compute, but by whether the system can override human control). Australia proposed an AI "kill switch" law requiring providers to maintain emergency shutdown capability. China's Supreme Court ruled cloning someone's voice or likeness without consent can violate their legal rights. And US intelligence agencies accused Chinese AI labs of systematically extracting knowledge from American frontier models. There's also a genuinely interesting thread across all of this: different countries disagree wildly on AI regulation in general, but there's real convergence forming around one specific thing: deepfakes and control over your own likeness. Check out the ep 1 (5 stories, first episode of a new weekly format): https://youtu.be/H54oiWU9WuY?si=sf0KB3KadKGkq3i3?utm_source=reddit&utm_medium=organic&utm_campaign=incident_series&utm_content=78-ep1-aigovweekly Question: Which of these actually has a shot at becoming real law: the superintelligence ban, or the kill-switch bill?