Open Data

JFrog extends DevSecOps playbook to AI governance | Computer Weekly

JFrog extends DevSecOps playbook to AI governance | Computer Weekly

JFrog is extending its DevSecOps platform to manage AI models with the same rigor applied to traditional software, treating them as analogous code artifacts. By unifying machine learning operations and governance into a single system of record, the company addresses the critical gap where mature software security practices have not yet extended to the less-regulated AI development environment. This approach ensures that AI projects benefit from established security standards rather than operating in isolated, vulnerable silos. Central to this initiative is the introduction of the Machine Learning Bill of Materials (ML-BOM), which tracks both model integrity and data provenance. This framework captures essential details regarding data sourcing, potential bias, and licensing, while incorporating governance principles to create a clear audit trail. Crucially, it allows organizations to flag closed-source models with unknown origins, enabling risk-based decisions for highly regulated industries that cannot afford to overlook privacy or ethical violations embedded within training datasets. This consolidation is vital for open data because it promotes transparency and accountability in the AI supply chain. By integrating with broader development ecosystems and preventing the proliferation of disconnected point solutions, JFrog helps eliminate blind spots that often obscure data lineage. As agentic AI adoption grows, this unified, open approach ensures that organizations can monitor and govern their AI assets securely, fostering trust in how data is used and processed across complex, interconnected systems.

Source: computerweekly.com
Published on 2025-09-03

Related news