A threat actor impersonated roughly 32 AI bots to scan enterprise web servers for credentials, cloud secrets, and private SSH ...

Anthropic wants California's attorney general empowered to seek court-ordered suspensions of dangerous models, while OpenAI favors mandatory shutdown controls enforced by developers.

A new phishing campaign disguises credential theft as an invitation to a private AI event, using GoTo Webinar branding, a ...

The board approved the AI rollout without you, so unless you lock down its MCP servers now, you’ll be explaining the breach ...

Verizon’s 2026 Data Breach Investigations Report — which is based on the combined evidence of multiple incident response ...

Attackers are bundling legitimate OpenAI and Anthropic executables with malicious DLLs, using SEO-poisoned sites and GitHub ...

Symantec researchers have captured instructions for a self-directed offensive agent that can run up to 100,000 steps, steal ...

Having backups isn’t the same as being able to recover, because after an attack, the hard question is what you can still ...

An out-of-bounds write in Google’s Brotli decoder affects multiple software ecosystems, prompting security updates for .NET, ...

Tasked with writing a rule to detect malicious software downloads, Qwen3.8-Flash-Next instead examined its test environment ...

A high-severity flaw could expose AI training data, model artifacts, and cloud credentials when a crafted model configuration ...

AI and ‘indexing-as-a-service’ are enabling cybercriminals to extend the blast radius of breaches by comprehensively ...