We curate threat intelligence to provide situational awareness and actionable insights
Threat Identifier Detections
Atomic detections that serve as the foundation of our detection framework.
Threat Scenario Detections
Risk, pattern, and sequence-based detections utilizing the outputs of Threat Identifiers as a means of identifying actual threats.
Reports Hot Off the Forge
• Threat News Reports
• Trending Threat Reports
• ResearchArticles
Forge Report: First Half Threat Trends of 2024
Featured Threat Reports
All Threat Reports
New LLM Jailbreak "Immersive World" Proves AI Can Be Manipulated for Malware Creation
Cato Networks' "Immersive World" jailbreak bypassed LLM safeguards, enabling AI models to create a working Chrome infostealer. Using narrative manipulation, even non-technical users could produce malware. The findings expose critical vulnerabilities in GenAI systems and highlight the rise of “zero-knowledge” threat actors empowered by accessible AI tools.
Europol Report Warns: AI-Powered Cybercrime Reshaping Global Criminal Networks
Europol’s 2025 EU-SOCTA report reveals how AI is transforming cybercrime, enabling deepfakes, phishing, and RaaS at scale. Criminal networks and state-sponsored actors exploit AI to destabilize economies, automate fraud, and blur attribution. Europol calls for urgent action to disrupt financial flows and counter AI-driven threats to global stability.
UAT-5918 Overlapping TTPs from Critical APT Groups, Attacks with An Arsenal of LOLBins & External Hacking Tools
Cisco Talos attributes attacks on Taiwan's critical infrastructure to UAT-5918, a threat group using LOLBins, web shells, and tools like FScan and Impacket. Overlapping TTPs with Volt Typhoon and Flax Typhoon raise attribution concerns. The group seeks persistence, credentials, and exfiltrates sensitive data via common attack frameworks.
SocGholish Malware Identified as Key Enabler of RansomHub Ransomware Attacks
Trend Micro has linked SocGholish malware to RansomHub ransomware deployments in a campaign known as Water Scylla. Delivered via fake browser updates and compromised sites, SocGholish enables initial access and credential theft. The U.S. faces the highest infection rates, with attackers leveraging SSH tunnels and browser credential theft.
Massachusetts Utility Targeted by Advanced Persistent Threat Linked to China
A Dragos investigation revealed VOLTZITE, tied to China’s Volt Typhoon, breached a Massachusetts utility in 2023, maintaining OT network access for over 300 days. The group exfiltrated critical grid data but avoided customer PII. LELWD has since enhanced network defenses and segmentation to prevent future intrusions.
Compromised Credentials Enabled PowerSchool Data Exfiltration, CrowdStrike Report Finds
A CrowdStrike-led investigation found that compromised credentials enabled threat actors to access PowerSchool SIS data via the PowerSource portal. Data from student and teacher records was exfiltrated. No malware or further system access was detected. PowerSchool enforced MFA, SSO, and VPN restrictions and offered identity protection to affected users.
About the Forge & Threat Reports
Our mission is to assess the operational behaviors of all threats to provide the community, and our customers, with actionable information and enterprise-ready detections in order to defend themselves in an ever- changing threat landscape.
Intelligence Levels for Threat Reports
Tactical
Detectable threat behaviors for response with threat scenarios or threat identifiers.
Strategic
General information security news, for awareness.
Whitepapers
The World's Best SOC Teams Use Anvilogic
.svg)
.svg)
.svg)
.png)
Build Detections You Want, Where You Want
