--- title: Python libraries in AI/ML models can be poisoned w metadata description: Security researchers discovered vulnerabilities in popular Python AI/ML libraries (NeMo, Uni2TS, FlexTok) from Nvidia, Salesforce, and Apple that allow remote... canonical: https://daily.dev/posts/python-libraries-in-ai-ml-models-can-be-poisoned-w-metadata-4tbuftlcz twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: Python libraries in AI/ML models can be poisoned w metadata | daily.dev og:description: Security researchers discovered vulnerabilities in popular Python AI/ML libraries (NeMo, Uni2TS, FlexTok) from Nvidia, Salesforce, and Apple that allow remote... og:url: https://daily.dev/posts/python-libraries-in-ai-ml-models-can-be-poisoned-w-metadata-4tbuftlcz og:image: https://api.daily.dev/og/posts/4tbUFTlCZ.png og:image:alt: Python libraries in AI/ML models can be poisoned w metadata og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # Python libraries in AI/ML models can be poisoned w metadata **[The Register](https://daily.dev/sources/theregister)** · 6 min read · 0 upvotes · 0 comments ## Summary Security researchers discovered vulnerabilities in popular Python AI/ML libraries (NeMo, Uni2TS, FlexTok) from Nvidia, Salesforce, and Apple that allow remote code execution through poisoned metadata. The flaws stem from unsafe use of Hydra's instantiate() function, which can execute arbitrary code when loading model configurations. Attackers could hide malicious code in metadata that executes automatically when files are loaded. The libraries, used in Hugging Face models with millions of downloads, have been patched with CVEs issued. Meta updated Hydra documentation with security warnings, though no in-the-wild exploitation has been observed. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Similar posts on daily.dev - [Remote Code Execution With Modern AI/ML Formats and Libraries](https://daily.dev/posts/remote-code-execution-with-modern-ai-ml-formats-and-libraries-kzzcz2ymw) · Unit 42 · 0 upvotes · 0 comments - [Hugging Face and ClawHub compromised with hundreds of malicious AI models and agent skills as supply chain attacks target AI infrastructure](https://daily.dev/posts/hugging-face-and-clawhub-compromised-with-hundreds-of-malicious-ai-models-and-agent-skills-as-supply-xjnujrk1y) · The Next Web · 4 upvotes · 0 comments - [Hugging Face Transformers RCE flaw enables stealthy compromise via AI model configs](https://daily.dev/posts/hugging-face-transformers-rce-flaw-enables-stealthy-compromise-via-ai-model-configs-uysdnnbz0) · CSO Online · 1 upvotes · 0 comments --- Tags: [#security](https://daily.dev/tags/security), [#machine-learning](https://daily.dev/tags/machine-learning), [#python](https://daily.dev/tags/python), [#vulnerability](https://daily.dev/tags/vulnerability) [View this post on daily.dev](https://daily.dev/posts/python-libraries-in-ai-ml-models-can-be-poisoned-w-metadata-4tbuftlcz) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"Python libraries in AI/ML models can be poisoned w metadata","url":"https://daily.dev/posts/python-libraries-in-ai-ml-models-can-be-poisoned-w-metadata-4tbuftlcz","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/python-libraries-in-ai-ml-models-can-be-poisoned-w-metadata-4tbuftlcz"},"datePublished":"2026-01-13T21:19:11.569Z","dateModified":"2026-01-13T21:19:33.105Z","description":"Security researchers discovered vulnerabilities in popular Python AI/ML libraries (NeMo, Uni2TS, FlexTok) from Nvidia, Salesforce, and Apple that allow remote...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/1d50b4255bdea8146aa7225f9619d231?_a=AQAEulh","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/1d50b4255bdea8146aa7225f9619d231?_a=AQAEulh","isAccessibleForFree":true,"articleSection":"The Register","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"The Register","logo":"https://media.daily.dev/image/upload/t_logo,f_auto/v1/logos/66aa2113fdad463992ffcbf0e8963fda","url":"https://daily.dev/sources/theregister"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/python-libraries-in-ai-ml-models-can-be-poisoned-w-metadata-4tbuftlcz","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":0},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"security,machine-learning,python,vulnerability","timeRequired":"PT6M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"The Register","item":"https://daily.dev/sources/theregister"},{"@type":"ListItem","position":3,"name":"Python libraries in AI/ML models can be poisoned w metadata"}]} ```