Can you explain which part of the AI Act killed what exactly? If you bother to read it, it's just common sense such as saying that a company can't just delegat

Can you explain which part of the AI Act killed what exactly? If you bother to read it, it's just common sense such as saying that a company can't just delegate all decision making to "AI" and hide behide it for critical things. For instance, in the US healthcare companies use "algorithm" as the excuse for letting people die, this would be illegal under the AI Act. That doesn't in any way impact any AI startups selling "AI", nor companies buying/doing AI. Hell, there's AI in healthcare in France already.