The supply chain attack campaign's shift toward developer endpoints and cloud access signals what could be next for the cybercrime outfit.

Mail bombing is getting a new look with a wide-scale, ongoing social engineering campaign that might be tied to the Black Basta ransomware gang.

A new AI agent prompt injection technique exploits how coding assistants consume local instruction files, blurring the line between text documents and executable actions.

Rising AI, social engineering, and cloud risks are pushing carriers beyond risk transfer and deeper into cybersecurity itself.

In this video, Dark Reading editors discuss some of the news they didn't get a chance to cover, from busting up North Korea's stateside infrastructure in "Operation Lonestar Justice" to Staples' gift card bonanza.

Security teams are scrambling to remediate two critical PHP vulnerabilities following leaked details and reports of exploitation attempts.

The perp, who has Saudi ties, was arrested for allegedly bugging an OpenAI conference room, highlighting a preventable insider threat vector.

Researchers say attackers authenticated into Microsoft 365 tenants using ConnectWise credentials; ConnectWise says customer impact remains unconfirmed.

As agents escape containment and access external systems, responsibility remains human. Accountability, however, is much less settled.

A sprawling cybercrime marketplace supplied cloud infrastructure and phishing kits that helped fuel more than 1 billion phishing messages worldwide.

Anthropic's latest study combines IoT and network tests to explore AI self-replication, which researchers said would be "challenging" to shut down.

Edward Dubrovsky, a longtime cybersecurity executive, was arrested and charged in Pennsylvania last week under conspiracy to extort money.

In this video conversation, Dark Reading editors discuss some of the news they didn't get a chance to cover, from the arrest of a suspected ShinyHunters operative to the compromise of a Pentagon-run data center.

As AI agents gain authority over business systems, attackers can manipulate them like business email compromise (BEC) victims.

Anthropic has merged Project Glasswing into a tiered access program for its advanced cyber LLMs, including Opus, Sonnet, and Mythos.

Threat actors are now hiding payloads by using DNS TXT records and browser cache pre-fetching, making it tougher to spot early attack stages.

Not quite an EDR-killer, but the proof-of-concept cyber technique creates a silent virus detection gap while service runs normally, no exploit required.

"Rogue AI" terminology anthropomorphizes LLMs and shifts risk responsibility from vendors. Defenders should treat agents as untrusted, nondeterministic software systems, not sentient beings with malicious intent.

In yet another ClickFix-style campaign, threat actors abuse legitimate domains from OpenAI and Google to fool unsuspecting users.

A patched Unsloth Studio vulnerability allows malicious AI models to execute arbitrary Python code during inspection, via the trust_remote_code setting.