Do people not know about abliteration?

A Reddit user argues that open-weight AI models cannot be aligned, making them an unviable alternative to proprietary models despite concerns about companies like OpenAI and Anthropic. The user dismisses the idea of China winning the AI "race" for better alignment. Instead, they propose that a government agency should regulate and license advanced AI models, similar to biotechnology, to ensure a stable future society.