A human brain is a glorified auto correct.

Fact is, the AI that is dangerous is not giving advice about how to get to a car wash. It’s building hacks and scams etc. At a level and scale we’ve never seen before.

I think the lesson that should be taken from this is not that we need to prevent open source AI from coming out. We should recognize that Anthropic is in control of weapons that should be regulated by the government more heavily than it is. And society as a whole should be preparing for the “mythos” style AI to be open sourced. Which is inevitable from the sounds of it.