AI Flaws in Amazon Bedrock, LangSmith, and SGLang Enable Data Exfiltration and RCE

Cybersecurity researchers have disclosed details of a new method for exfiltrating sensitive data from artificial intelligence (AI) code execution environments using domain name system (DNS) queries.
In a report published Monday, BeyondTrust revealed that Amazon Bedrock AgentCore Code Interpreter’s sandbox mode permits outbound DNS queries that an attacker can exploit to enable interactive shells

Source

What's Hot

Critical Unpatched Telnetd Flaw (CVE-2026-32746) Enables Unauthenticated Root RCE

Researchers Sound the Alarm on Vulnerabilities in AI-Generated Code

Claude Code Security and Magecart: Getting the Threat Model Right

Critical Unpatched Telnetd Flaw (CVE-2026-32746) Enables Unauthenticated Root RCE

Researchers Sound the Alarm on Vulnerabilities in AI-Generated Code

Claude Code Security and Magecart: Getting the Threat Model Right

North Korean Hackers Turn JSON Services into Covert Malware Delivery Channels

macOS Stealer Campaign Uses “Cracked” App Lures to Bypass Apple Securi

Cyber M&A Roundup: Cyber Giants Strengthen AI Security Offerings

Malicious Nx Packages in ‘s1ngularity’ Attack Leaked 2,349 GitHub, Cloud, and AI Credentials

Near-ultrasonic attacks on voice assistants

Most Popular

North Korean Hackers Turn JSON Services into Covert Malware Delivery Channels

macOS Stealer Campaign Uses “Cracked” App Lures to Bypass Apple Securi

Cyber M&A Roundup: Cyber Giants Strengthen AI Security Offerings

Our Picks

Look out for phony verification pages spreading malware

Why you should never pay to get paid

What it is and how to protect yourself

Subscribe to Updates

What's Hot

AI Flaws in Amazon Bedrock, LangSmith, and SGLang Enable Data Exfiltration and RCE

Related Posts

Subscribe to Updates