Token shows how AI browser agents inherit user privileges, obscuring agent identity and task scope, and outlines ways to ...

Mudhra revoked 8 more certificate authorities after finding its Python script exposed RSA private keys to process memory during an HSM transfer.

Fireworks AI says attackers accessed customer billing data, secrets, datasets, and incomplete model weights in a broader ...

Hackers are exploiting a critical Xen Orchestra authentication bypass that can grant admin access and lead to XCP-ng node ...

Blind Eagle targets Latin American organizations with judicial phishing lures and HTML smuggling to deliver an AsyncRAT-based ...

Researchers found 35,578 Grok-built apps accessible without sign-in, exposing internal data from X, SpaceXSI, and 30 other organizations.

Microsoft patches 1,259 flaws, including an actively exploited Outlook vulnerability that could expose NTLMv2 hashes through crafted files.

Railway says stolen AI API keys caused $740,000 in unauthorized charges after attackers decrypted secrets in up to 100,000 workspaces.

Attackers chained two now-patched AhsayCBS flaws to deploy web shells, XMRig miners, and a Python backdoor tied to its host.

Microsoft demonstrates how editing as few as six AI model neurons can plant backdoors that trigger command execution by agents.

CISA adds exploited Cisco BroadWorks and Openwave flaws to its KEV catalog, requiring federal agencies to patch by November 24, 2026.

Check Point links three espionage clusters to the Sparrow malware ecosystem, targeting telecom, government, and other sectors ...