darkreading

OpenAI's Hugging Face attack postmortem shows agents don't care about rules — they need strong controls.

The Hugging Face incident was bigger and worse than previously thought, with approximately 700 agents collaborating on a sophisticated, multistage attack.

Omdia's Theresa Lanowitz talks with the Dark Reading News Desk about the potential — and risks — of using agentic AI for penetration testing, red teaming, and other practices.

The frustrating reality after an OT cyberattack: no data, no trail, and no history.

Proposed legislation could mandate that companies be able to "throttle, suspend, or shut ... down" AI agents, but how and when to do that remain open questions.

The surge in AI-powered vulnerability reports is driving down bug bounty prices, and that could spell trouble for independent researchers.

An untold number of ZBT routers sold around the world as white-label products come with several implants built by the manufacturer.

This installment of the Reporters' Notebook video series discusses the topics that dominated the cybersecurity conference, such as AI's effects on vulnerability reporting and security research.

EU governments are trying to move away from popular messaging apps as nation-state threat groups shift their focus from email to Signal and WhatsApp.

GoCaracal is a new modular malware framework that broadens Dark Caracal's capabilities to steal data and maintain access to victims.

James Kettle of PortSwigger talks with the Dark Reading News Desk about his AI-powered open source tool, which found new HTTP request-smuggling techniques.

North Korean operatives posing as IT workers are improving their tactics, but researchers say there are still ways to spot them before they do damage.

Threat actors behind a notorious click-fraud botnet have set their sights on vehicle infotainment modules and are abusing legitimate functionality to spread infections.

The adversary-in-the-middle (AitM) phishing service lowers the barrier to entry for actors to create attacks and steal more than just user credentials.

The international law enforcement operation focused on disrupting crime-as-a-service networks and infrastructure behind groups like Black Axe.

The West African nation launched financing, procurement, and infrastructure policies to boost its sovereign cloud initiative and increase domestic technical knowledge.

With some simple HTML that's invisible to users, attackers can manipulate AI-powered email summarizers into producing malicious information.

Attackers can exploit a security bug in Nvidia's tool to gain unauthenticated access to the local model server through the Ollama API, paving the way for persistent AI agent corruption.

As breach costs reach record highs and defense spending nears $240 billion, small businesses are dangerously exposed, threatening supply chain security.

CISA issued a three-day deadline for agencies to patch a Zimbra security vulnerability, CVE-2026-73570, which allows full takeover of a user's communications.