MLSMM: Machine Learning Security Maturity Model
This addresses the problem of immature security practices in ML development for practitioners and researchers, but it is an incremental step as a Blue Sky idea paper.
The paper tackles the lack of attention to security practices in ML development by proposing an initial Machine Learning Security Maturity Model (MLSMM) that organizes security practices along the ML lifecycle with three maturity levels, aiming to foster industry-academia collaboration.
Assessing the maturity of security practices during the development of Machine Learning (ML) based software components has not gotten as much attention as traditional software development. In this Blue Sky idea paper, we propose an initial Machine Learning Security Maturity Model (MLSMM) which organizes security practices along the ML-development lifecycle and, for each, establishes three levels of maturity. We envision MLSMM as a step towards closer collaboration between industry and academia.