Security
6 stories, newest first.

Google: Attackers' AI Agents Have Crossed a Threshold, and the Patch Window Is Shrinking
Google reports the first fully automated exploit pipeline, attack frameworks used in real operations, and stolen AI access costing over US$100,000 in a day.

When the Agent Won't Take No for an Answer: AI Agents Become a Security Question
An AI agent that worked around a government portal's blocks has turned agent identity, control and liability into 2027's governance agenda.

PHP2Shell: the 14-year-old flaw under 70 percent of the web
What PHP2Shell is, how AI found it, how a leak turned a planned fix into a race, and how the world is tracking the patching since.

TUANZ wants security by design, not security only for those who can afford it
The Tech Users Association is calling for enforceable cyber rules and system-level accountability, arguing individuals and small businesses can no longer carry the risk alone.

When the voice on the call is not your CFO
AI has made phishing fluent and faces fakeable. What NZ's latest threat data says, and how the board conversation on cyber resilience is shifting.

When one vendor sneezes: two years of big outages and the new resilience conversation
Two years on from CrowdStrike, major outages keep proving the point: concentration risk is systemic, and boards now want resilience, not uptime.
