Software supply chain company JFrog Ltd. today introduced machine learning model management capabilities designed to streamline the management and security of machine learning models.

Currently in beta test mode, the new support adds additional capabilities to the JFrog Platform to bring machine learning model delivery in line with an organization’s existing DevOps practices and to accelerate and govern the release of machine learning components.

JFrog argues that as artificial intelligence and machine learning use continue to grow, users often face challenges such as cost, lack of automation, lack of expertise and inability to scale up. JFrog’s new ML Model Management has been designed to address those issues.

New support features include allowing users to proxy the popular public machine learning repository Hugging Face to cache open-source AI models companies rely on from deletion or modification. Users can scan machine learning model licenses to ensure compliance with company policies, detect and block the use of malicious models, store homegrown or internally augmented models with robust access controls and versioning history for greater transparency, and bundle and distribute models as part of any software release.

Also announced today were new MLOps capabilities in the JFrog Platform that are said to infuse security at the binary level in every stage of the software development lifecycle to ensure applications are traceable, reliable, compliant and secure.

The new capabilities in the JFrog platform include static application security testing and open-source software cataloging, in addition to the enhanced machine learning model security capabilities.

“Increasing numbers of organizations are starting to incorporate ML models into their applications and with several government regulations requiring software vendors to list exactly what’s inside their software, we believe it won’t be long before these guidelines grow to include ML and AI models as well,” Yoav Landman, co-founder and chief technology officer of JFrog, said ahead of the release.

JFrog was last in the news in July when it launched JFrog Curation. The automated DevSecOps solution checks and blocks infected open-source or third-party software packages and their dependencies before they enter a company’s development pipeline.

Image: JFrog