How do you know who to trust?

I've been actively studying trust for the better part of a decade, but, recently, as I have been digging through the published research, most of the thinking about who to trust becomes a conversation about what to trust. It all seems to turn toward human-computer interface questions or ecommerce, and, of course,  AI. Even the piece that I am currently co-authoring is HCI related. 

I'd like to consider this as a cross-discipline concept, sociological, anthropological, psychological, etc., so... How do you know who you can trust? What are the indicators or signals? What happens when trust erodes?