Sophos Urges Rules at Sensitive Uploads Before AI Judges Intent
Endpoint AI can flag unusual access, but a familiar-looking transfer may still be harmful. Put explicit policy at the upload and measure the review queue.
Read article →AI
AI systems, model security, agent abuse, and attacks that use generative tools.
Latest in AI
Microsoft and NVIDIA Map Open-Weight Model Risk to Enterprise Deployments
The joint guidance separates model evaluations from the permissions, data access and tools of the application that uses a model.
Read article →
Volexity Volcano Recovers Browser Exploit Traces After Cleanup
A replayed Chrome-to-Windows intrusion left encoded script in free memory. Capture RAM early, then use host artifacts to guide wider collection.
Read article →
CL-UNK-1211 Plants PHP Web Shells on Government Linux Servers
Attackers changed PHP configuration to reload a web shell, stole passwords through a sudo alias and searched government mailboxes for social media access.
Read article →
Bisheng SSO Issues Admin Tokens Without Checking the Caller
The legacy SSO route issues a JWT for a supplied username and remains unpatched; separate cross-user assistant and workflow flaws have release-specific fixes.
Read article →
Fake event invites target Taiwan researchers’ Google accounts
Talos traced event invitations and altered poster QR codes to a live Google sign-in relay. Inspect the actual registration destination and review affected accounts.
Read article →
Apex Accelerator's fake audits deliver malware through npm tools
Two Apex npm tools fetch executables and attacker-selected commands; one also sends coding-agent conversations to an Apex server. Find the tools and revoke exposed secrets.
Read article →
Cisco Talos Measures Malware Prompts That Mislead AI Scanners
Plaintext instructions in malware samples shifted some model triage toward benign. Search extracted strings and keep sample text out of the scanner's instruction channel.
Read article →
Microsoft FORGE Finds 140 Windows CVEs, Submits 155 Open-Source Reports
Microsoft's research shows how bug volume can overwhelm validation and release work. Check affected dependency builds and carry reproducible findings through a shipped fix.
Read article →
Microsoft urges 24-hour fixes for critical on-premises systems
Microsoft urges faster fixes for critical on-premises systems. Check patch deployment logs, installed-update status and Baseline Security Mode configuration exceptions.
Read article →
Cisco Talos maps RMM, decoys and agent controls to disrupt attackers
A legitimate remote-support agent can give a ransomware operator persistent access. Inventory approved tools, watch credential-access behavior and make agent sessions stoppable.
Read article →
Cisco Talos Uses CAIRN to Hunt AI Malware From Sample Metadata
CAIRN groups malware candidates by AI-related metadata, then separates artifact presence from behavior and confirmed families. Triage noisy matches before calling an incident.
Read article →
Microsoft Brings Sensitive AI Upload Blocking to Entra Networks
Purview classifies files and text while Entra enforces network policy for employee and on-behalf-of agent traffic. Confirm forwarding and an active blocking rule before relying on it.
Read article →Showing 1–13 of 47