An open letter signed by dozens of security experts asked the government to reverse export restrictions on Anthropic's Claude Fable 5 and Mythos 5 models.

The critical, three-stage attack is now patched, but it's part of a new group of AI prompt-injection issues that use hidden URLs and other variables.

Stay cool: Mythos 5 is an upgrade over Mythos Preview while Fable 5 is Mythos "made safe for general use," Anthropic explains.

Security research inadvertently led organizations to believe they were being breached through their ServiceNow instances.

“Ghost-Sender" is the result of a widespread misconfiguration, according to researchers, and evidence indicates it's being actively abused in the wild.

A newly discovered, critical zero-day vulnerability is under attack; a Qilin ransomware affiliate has been blamed for at least one incident.

The White House's executive order establishes voluntary framework for early government access to frontier models while investing in federal security.

Python scripts were used to test malware against endpoint detection and response agents from Sophos, CrowdStrike, and Windows Defender.

A prompt injection flaw in Google Gemini's voice assistant let attackers hide malicious commands in notifications, enabling social engineering and more.

China is stealing data from high-value targets via a sneaky, double-layer spear-phishing campaign that includes the Azureveil malware.