As organizations increasingly rely on machine learning for critical decision-making, ensuring the security of ML systems becomes paramount. MLSecOps is a vital discipline that integrates SecOps with ML operations to protect models, data and the ML infrastructure. Just like DevSecOps, it upholds privacy regulations and compliance standards, addresses vulnerabilities, detects adversarial attacks and protects against data breaches, maintaining the integrity and confidentiality of models and data.
Very often, data scientists and ML engineers do not have enough proficiency or motivation to protect ML data and infrastructure, while security professionals are struggling to understand new complicated terminology. This presentation aims to close the skills gap for people who have the practical experience of defining and implementing DevSecOps and SSDLC practices.
Learning Objectives:
Understand the key concepts of MLOps that are relevant for security professionals.
Apply DevSecOps practices established in the organization to the MLSecOps roadmap.
Understand the gaps between DevSecOps and MLSecOps.
