Salemicus comments on Musk on AGI Timeframes

[Salemicus]

Humans can be extremely dangerous. Why wouldn’t a human-level AI be?