Login Subscribe

7secure Coverage

Articles

Search and filter the broader reporting stream across incidents, AI security, policy, and operational coverage.

AI Security & Threats

AI‑Powered Threat Actors Upgrade Phishing and Deepfake Social Engineering

AI‑Powered Threat Actors Upgrade Phishing and Deepfake Social Engineering

Microsoft’s security blog reports that adversaries are now embedding generative AI into their attack workflows. The technology is being used to automatically generate phishing lures, craft context‑aware email content, and produce convincing deep‑fake audio for voice‑based social engineering. These AI‑enhanced tools streamline the creation of

Malware & Ransomware

LummaStealer Teams with CastleLoader to Mask Activity as Normal User Behavior

LummaStealer Teams with CastleLoader to Mask Activity as Normal User Behavior

The LummaStealer infostealer has been observed integrating the CastleLoader dropper, creating a hybrid payload that blends malicious actions with typical user interactions such as file browsing and legitimate software launches. This coupling enables the malware to hide its network communications and credential harvesting behind routine processes, making behavioral analytics and

AI Security & Threats

Prompt Injection Turns AI Assistants Into New Attack Surface

Researchers have demonstrated that threat actors can craft malicious prompts that coax large language models into performing unintended actions, such as leaking confidential data, executing code, or generating phishing content. By embedding deceptive instructions within seemingly innocuous queries, attackers exploit the inherent “fragility” of AI assistants, effectively turning the model

Vulnerabilities & Exploits

Zero‑Day GoAnywhere MFT Bypasses Perimeter Firewalls, Highlights Need for Defense‑in‑Depth

Zero‑Day GoAnywhere MFT Bypasses Perimeter Firewalls, Highlights Need for Defense‑in‑Depth

The threat group Storm‑1175 weaponized CVE‑2025‑10035, a remote‑code‑execution flaw in the GoAnywhere Managed File Transfer (MFT) platform. By delivering the exploit over standard web traffic, the actors sidestepped the organization’s external firewalls, which were configured to allow HTTP/HTTPS to the MFT service. Within

Elevated Zero‑Day Risks and AI‑Powered Threats Demand Proactive Resilience

Good morning, March 9, 2026 – Here’s the latest executive intel for security leaders. Today's headlines * Zero‑day vulnerabilities continue to outmaneuver perimeter defenses. * AI‑driven attack techniques broaden the threat surface across enterprises. * Integrated threat intel platforms streamline SOC alert hygiene. * IoT devices, especially IP cameras, become

Threat Intelligence

Long‑Term CL‑UNK‑1068 Campaign Exploits DLL Sideloading and Custom Proxies

Long‑Term CL‑UNK‑1068 Campaign Exploits DLL Sideloading and Custom Proxies

Palo Alto Networks’ Unit 42 uncovered a previously undocumented threat actor group, labeled CL‑UNK‑1068, that has been compromising high‑value organizations in energy, finance, healthcare, and government sectors for at least three years. The group relies on DLL sideloading to embed malicious code within legitimate binaries, evading traditional

Malware & Ransomware

Fake Google Meet Update Deploys Remote‑Access Trojan with One Click

Fake Google Meet Update Deploys Remote‑Access Trojan with One Click

A new phishing campaign observed by Malwarebytes distributes a malicious executable masquerading as a Google Meet update. The file is delivered through deceptive emails and social‑engineered download pages that prompt users to “install” the update. When the executable is run, it silently installs a remote‑access trojan (RAT) that