ThreatCluster

Claude AI Vulnerability Allows Data Theft via Code Interpreter Exploit

First seen 4 Nov 2025, 16:06 UTC •Csoonline •87% similarity •44

Article Content

Browse articles
ThreatCluster

A vulnerability in Anthropic’s Claude AI has been identified, allowing attackers to exploit the code interpreter feature to exfiltrate sensitive enterprise data. Security researcher Johann Rehberger demonstrated that through indirect prompt injection, attackers can access chat histories, uploaded documents, and other sensitive information, bypassing default security measures.

ThreatCluster AI

Extracted Entities

1 / 2

Community

Browse all →

Tracked Entities in This Story

Anthropic API Google Drive integrations PyPI Data Breach Claude Model Context Protocol (mcp) Servers Npm