What does it take to turn a tool into a talking tool and that into an ultimate authority? Generative artificial intelligence (GenAI) in its diverse forms, such as large language models (LLMs), is celebrated as a useful tool. But LLM‐based conversational agents, or chatbots, the software applications through which ordinary users are likely to engage with LLMs on a regular basis, can seem, for many, to harbour uncanny insights or transcendental knowledge whose sources are inexplicable – resembling what Marshall Sahlins called ‘metahumans’. The temptation to treat chatbots as metahuman is just an extreme case of a more general process. The inferences that chatbot users draw about GenAI's human or superhuman properties arise in part from the apps' role as an interlocutor in social interaction. Their authority draws on the pragmatics and semiotics of speech. Much as a prophet embodies and legitimates the authority of divinity, so chatbots can endow the profit‐seeking strategies of their corporate owners with a supernatural aura. That such effects can seem intuitively real is, in part, because humans and metahumans address one another on semiotically unequal grounds.
Webb Keane (Fri,) studied this question.