> if states can pass restrictive laws on AI in absence of a correspondingly negative motivating event...

If you mean besides the extensive harm to air quality, the large land fingerprint of data centers, the massive strain on water resources and treatment facilities, the insane electricity demands resulting in skyrocketing prices pushed onto everyone else, the deafening noise pollution, and what they've done to the price of RAM, then sure. And that's just the data centers!

The usage of AI itself has resulted in all kinds of harm and even actual deaths. AI has wrongfully denied people healthcare coverage they were entitled to preventing or delaying needed surgeries and treatments. There's a growing list of LLM related suicides (https://en.wikipedia.org/wiki/Deaths_linked_to_chatbots). The use of AI in parole systems has kept people locked behind bars when they shouldn't have been due to biases in the bots making decisions. AI used for self-driving driving cars have killed pedestrians and other drivers. There are thousands of AI generated harms tracked here: https://airisk.mit.edu/ai-incident-tracker