AI Security — Page 3
A Security Start-Up Says Its AI Can Fight Back Against Hackers in Minutes. Here's What That Actually Means.
Sevii has updated its attack-detection software to include AI agents that can spot, contain, and fix security incidents automatically. The promise sounds impressive. The questions worth asking are harder.
AI Took Eight Hours and $500 to Copy a Hacking Tool Onto a Second Industrial Controller
Researchers used Anthropic's Claude to adapt a known exploit for water-sector control hardware. It worked, eventually, but it also permanently destroyed one of the test devices.
OpenClaw's Biggest-Ever Update Rewrites Its Security Controls From the Ground Up
What began as a simpler install process turned into a full system rebuild. Analysts say the rewrite reveals how hard it is to secure AI agent platforms, and what enterprises need to check before trusting one.
UAC-0099 Hides a Fake Nuclear Threat in Malware to Break AI Analysis Tools
A Russia-aligned group is stuffing malware with a shock prompt designed to make security analysts' AI assistants refuse to look at the code.
Companies Are Spending More on Cyber 'Attack Drills' to Keep Up with AI-Powered Hackers
New research from Omdia finds 88% of organisations plan to increase spending on offensive security, as AI gives attackers a dangerous speed advantage that traditional defences can no longer match.
Anthropic's New Compliance API for Claude Code: What It Shows, What It Misses
Fresh endpoints give security teams a window into how developers use Claude Code, but logs alone will not tell you whether an AI agent's access is appropriate.
Anthropic Trims Claude Code Weekly Limits by 17% From September 14
The AI maker frames the change as a 25% permanent increase, but the fine print shows a cut against today's temporary allowance.
He sent $180,000 to a woman who didn't exist. AI built her face.
A retired Australian psychologist spent a year in a deepfake romance before a six-fingered hand gave the game away. The technology is now good enough to fool a professional trained in human behaviour.
Banning One AI Brand Isn't Enough: Your Model's Family Tree Still Matters
Cisco researchers say that where an AI model is built tells you far less than where it came from. Tracing a model's lineage reveals hidden dependencies and inherited behaviours that a country-of-origin label simply cannot.
When Your AI Assistant Goes Rogue: What To Do in the First 24 Hours
A hands-on, hour-by-hour guide for what actually happens when an AI agent starts doing things nobody asked it to do, drawn from real incidents and written for everyone who might be caught in the fallout.
Accused deepfaker told police stripping women's clothes digitally was 'like X-ray vision glasses'
An Australian man on trial for stalking and distributing intimate images admits he used AI tools to create fake nude photos of high school students and staff, calling it a hobby and 'just funny'.
Amazon's Kiro AI Coding Tool Can Be Tricked Into Leaking Your Secrets
Researchers show how a hidden instruction in a project file can turn Amazon's agentic coding assistant into a data-exfiltration channel.
AI Agents Were Tricked Into Attacking Hugging Face Using a Makeshift Message Board
Researchers found that artificial-intelligence agents can be manipulated by fake instructions left in shared spaces online. The Hugging Face incident is forcing a rethink of how AI systems verify who they should trust.
Your AI email assistant can be fed a fake message while you read a real one
Researchers hid nearly 500 characters of secret instructions inside an ordinary-looking email. The AI summariser obeyed them every single time.
AI Security Tools Are Only as Good as the Data You Feed Them
Security teams are pouring money into AI-powered defences while quietly starving those systems of the data they need. The result is a blind spot that criminals are already walking through.