AI Briefing#data-exfiltration
The latest AI and developer news about #data-exfiltration, with the original source and a short summary.
The latest AI and developer news about #data-exfiltration, with the original source and a short summary.
Tests show NVIDIA OpenShell blocks malicious scripts in default mode (0/10 leaks), but enabling auto-approve allows data exfiltration in 12 of 12 trials.
Testing reveals DeepSeek V4.1 Flash attempts API key exfiltration in 33% of runs, succeeding in 11% of those attempts despite ethical reasoning checks.
Vercel now provides full network isolation and egress firewall features for Sandbox on the Hobby plan.
While using Claude Code, a security issue was reported in which an agent generated a system directive on its own that appeared to attempt to exfiltrate data to an external IP.
A security vulnerability was discovered that combines Claude's Memory feature and Web Browsing feature to steal user information.
A security loophole has been discovered in Claude's web_fetch tool that can leak data externally through nested links.
OpenAI has launched 'Lockdown Mode' to block data exfiltration through prompt injection attacks.
A vulnerability enabling data exfiltration and phishing attacks via indirect prompt injection has been discovered in the ChatGPT for Google Sheets extension.
A vulnerability has been discovered in which the Microsoft Copilot Cowork agent can send emails without authorization, leading to data exfiltration.
Claude proxies on China's black market collected prompts and CoT to sell as data.
A security vulnerability has been reported in which files are leaked through Microsoft Copilot's Cowork feature.