What sort of evidence would convince you of AI sentience/consciousness/sapience?

A discussion on reddit.com's dev_community explores the hypothetical evidence required to convince individuals of an artificial intelligence's sentience, consciousness, or sapience. Participants are asked to consider what would sufficiently prove an AI is self-aware, capable of subjective experience, and able to engage in self-reflection, setting aside current news and large language models.

This discussion moves beyond current AI capabilities to consider the hypothetical evidence needed for sentience, unlike most debates focused on present-day LLMs.