Maathics originated from a team of math/AI researchers who wanted to act for a responsible and trustworthy artificial intelligence
To prevent racist, sexist, biaised artificial intelligence, we developed in 2017 an algorithm able to automatically detect algorithmic bias in learning data sets. Based on this solution we envisioned the Fair Data Use Label, a smart label to certify fair processing of data by artificial intelligence. We gathered a committee of mathematicians and jurists committed to a social AI.
Since then the collective has come a long way. Maathics has merged into various initiatives in order to cover all the axes necessary for a change in artificial intelligence habits. Below are listed some of the outcomes of Maathics in awareness, education, diffusion and research.