"ML Model Repositories: The Next Big Supply Chain Attack Target"

Repositories for Machine Learning (ML) models, such as Hugging Face, provide threat actors with the same opportunities to sneak malicious code into development environments as open source public repositories. In a presentation titled "Confused Learning: Supply Chain Attacks through Machine Learning Models," two Dropbox researchers will demonstrate multiple techniques that threat actors can use to distribute malware through ML models on Hugging Face. The methods are similar to those that attackers have successfully used to upload malware to open source code repositories, emphasizing the importance of organizations implementing controls to thoroughly inspect ML models. This article continues to discuss ML model repositories becoming a supply chain attack target.

Dark Reading reports "ML Model Repositories: The Next Big Supply Chain Attack Target"

Submitted by grigby1