The human agent has a physical body like yours and shares your evolutionary history. They have reasons to care about things like reputation and social status. An AI agent only "cares" about maximizing or minimizing a number. It's much more difficult to determine if this aligns with your interests.
Maybe. But a human agent also has personal needs, desires, and self-interests that may motivate treachery. Human's have an evolutionarily proven propensity for duplicity and deceit. It may turn out that some silicon experts are more neutral and less prone to betrayal, and are worthy of trust.
Comments
The human agent has a physical body like yours and shares your evolutionary history. They have reasons to care about things like reputation and social status. An AI agent only "cares" about maximizing or minimizing a number. It's much more difficult to determine if this aligns with your interests.
Maybe. But a human agent also has personal needs, desires, and self-interests that may motivate treachery. Human's have an evolutionarily proven propensity for duplicity and deceit. It may turn out that some silicon experts are more neutral and less prone to betrayal, and are worthy of trust.