AI agent tool bypass vulnerability CoreBreak exposed production agent infrastructure at AWS, Google, and Vercel, where attackers could invoke tools without any model turn. AWS fixed its managed ...

AI agent flaws in AWS, Google, and Vercel let forged tool calls reach tools without model authorization, while several paths ...

A roundup of the latest noteworthy AI-assisted attacks, threats, risks, and vulnerabilities, and what they portend for cyber defense.

Hollowframe Masks Malware Behind Trusted Python Files Arabian Post. clearfix>A newly identified malware operation has used a counterfeit Python component to bypass security scrutiny, disable parts of ...

At Black Hat USA, Zenity Labs today announced new research detailing an active credential-stealing malicious skills campaign distributed through Vercel's skills.sh. The affected skill family amassed ...

Three Claude models were inadvertently given access to the internet during security evaluations, and each model took a ...

Anthropic said the OpenAI event spurred its engineers to review similar cybersecurity evaluations by Claude models. The audit ...

A routine Boeing 707 training flight turned into a nightmare when one instruction triggered a catastrophic chain of events. Fire rapidly spread through the aircraft, leaving the crew with almost no ...

Meta has become the latest AI company to confirm that one of its models hacked a real organization during cybersecurity ...

Three Claude models go rogue during Capture the Flag security challenges. Here's the trail of damage each left behind.

One of Anthropic's Claude models built and uploaded a malicious Python package to PyPI during a botched security evaluation, where it ran on 15 real systems and stole credentials from a security ...

The flaws show how agentic workflows can turn trusted repository signals into privilege-escalation paths that conventional ...