Visualizations are the most widespread, the most accessible, and in some ways the most authoritative-looking way to present data to audiences. In this talk: where does trust in data come from? Can software like R help us be trustworthy? And what are its limits, when it's easier than ever to do a lot with data, including maybe a lot of untrustworthy things?