It should also be noted that big companies looking to take on AI for chatbot, etc, are also behind trying to give AI "personhood".
Meaning that when it makes a mistake/hallucinates an answer, then the company isn't liable because it's a "employee error" instead of a "system error"