Dr. Dan Villarreal has published a new article in Linguistics Vanguard, "Sociolinguistic auto-coding has fairness problems too: measuring and mitigating bias", with a companion GitHub repository. Building on earlier work on sociolinguistic auto-coding (SLAC) by Dr. Villarreal and others, this article demonstrates that, much like other predictive algorithms, SLAC is prone to biased predictions (e.g., making disproportionate predictions for female vs. male speakers). The article also describes numerous strategies for reducing bias, and the GitHub repository demonstrates how to implement these strategies in R code. Dr. Villarreal argues that since SLAC is relatively new in its development and use, "it is not too late to reverse the typical 'unleash the algorithm first, ask questions later' pattern".