With something like this, we basically have elevation of current and near-term AI to basically ITAR levels - the maximum penalties for ITAR violations is also

With something like this, we basically have elevation of current and near-term AI to basically ITAR levels - the maximum penalties for ITAR violations is also up to 20 years in jail. The challenge with these bans is that well... PRC has shown themselves to be pretty competent at pilfering ITAR/Top Secret information, say nothing about merely "company sensitive". . While on one hand, it's a perhaps a shrewd admission that no one really knows how to evaluate these AI models, and therefore we have no idea how to measure or contain PRC sourced AI models for potential negative impacts, most likely this will also setup a distinctly asymmetric information environment. PRC seemingly has a habit of pulling "bad" (for the US) surprises out of situations.