Many scientific papers and results are based on crowdsourced data. When assessing the quality, generality, bias, and ethical concerns, very little attention is paid to the specific ways in which the data was collected. However, these issues have a huge impact on the extent to which we can trust the results of the studies. In this talk, we first describe the general problem of algorithmic and AI bias and some solutions that have been identified. We then zoom in on crowdsourcing and provide a specialized discussion of the bias, ethics, and reliability challenges, as well as their proposed solutions. We conclude the talk with a call for a joint effort on reliable crowdsourcing.