Scworld Python Libraries in AI/ML Models Vulnerable to Metadata Poisoning
Article Content
Browse articles 
Vulnerabilities have been identified in popular AI and ML Python libraries used in Hugging Face models, allowing remote attackers to embed malicious code in metadata. This code executes automatically when a file with the poisoned metadata is loaded. The affected libraries include NeMo, Uni2TS, and FlexTok, developed by Nvidia, Salesforce, and Apple in collaboration with EPFL VILAB.
Ask AI about this cluster
Answers cite the sources they use
Updated 228d ago How this analysis works
More articles in this cluster (3)
Following this threat?
Track Hydra, Palo Alto Networks and CVE-2025-23304 in your own feed — you're alerted when they show up in new reporting, leak sites or exploitation.
Free account · no card needed
Continue Reading
Critical BeyondTrust 9.3 Vulnerability Actively Exploited by Threat Actor A critical vulnerability (CVE-2026-102764) in BeyondTrust Remote Support has been discovered and is actively being exploited by the threat group ARMORY CASTLE, reportedly linked to China. The vulnerability, which has a CVSS score of 9.3, allows unauthenticated attackers to upload files and potentially compromise host… 
PHP2Shell Vulnerabilities Exploited by AI-Driven Attackers Threat actors have exploited PHP2Shell vulnerabilities (CVE-2026-109618 and CVE-2026-109619) to target a government entity in Europe, gaining access to sensitive data within two hours of the initial breach. The attackers utilized AI tools for reconnaissance and post-exploitation activities, including accessing…