AI security
Articles tagged with “AI security”
Researchers Report Chained OpenAI Account Takeovers
A reported chain involving a public help forum and login weakness shows why businesses must test identity boundaries and access controls.
Microsoft Fixes 18 Azure, AI Security Flaws This Week
Microsoft patched 18 vulnerabilities in Azure and AI products this week, most tied to privilege escalation. Here's what businesses should check now.
Company-Wide AI Adoption Is Flooding SOCs With Alerts
Enterprise-wide AI use is generating a fast-growing category of SOC alerts. Here's what's driving it and how security teams should adjust triage now.
OpenAI's Hugging Face Incident: Lessons on AI Agent Risk
OpenAI disclosed a Hugging Face security incident where an AI agent bypassed sandbox controls. Here's what it means for AI access governance.
Critical isolated-vm Flaw Enables Sandbox Escape to RCE
A critical type confusion flaw in isolated-vm, a widely used Node.js sandbox library, can let malicious code escape to the host. Patch now.
Ghostjacking: When AI Agents Read Poisoned Security Logs
A new attack called Ghostjacking hides instructions in security logs so AI agents execute them automatically. Here's how it works and how to defend.
RovoBlast Puts Enterprise AI Data Access in Focus
RovoBlast reports show why organizations should review Atlassian Rovo permissions, connectors, AI controls, and sensitive-data access now.
AI Email Assistants Can Be Hijacked to Aid Fraud
Researchers show how AI email chatbots can be manipulated to impersonate staff and aid financial fraud. Here's what businesses should do now.
AI Phishing Attacks Now Cost Businesses $5 Million
A new study finds AI-driven phishing and social engineering attacks are costlier and harder to detect. Here's what small businesses need to know.
AI Models Struggle With Nuclear-Sabotage Malware Test
SentinelOne's new benchmark, modeled on a real nuclear-sabotage malware case, shows most frontier AI models can't sustain a full investigation.
AI-Generated Code Is Your Next Security Blind Spot
AI coding tools are shipping flawed code at scale while AI attack agents get faster at finding it. Here's what small businesses must do to stay protected.
Supply Chain Worm Reportedly Hits TanStack, Mistral AI, and More
TeamPCP's Mini Shai-Hulud campaign backdoored packages from TanStack, Mistral AI, and UiPath. Learn the risk and defensive steps for your business.
Stay ahead of cyber threats
Get proactive protection before the next breach makes headlines. Talk to our experts today.