Urgent.News

What's breaking now, across thousands of outlets.

AI

AI-driven cyber threats surge as Australia reports 80% rise in security warnings

ARTIFICIAL intelligence is accelerating cyberattacks against government networks, critical infrastructure and businesses, Australia’s cybersecurity agency warned on Tuesday, reporting an 80% increase in alerts over the past year. The Australian Signa...

Australia's cybersecurity agency has issued a stark warning that artificial intelligence is accelerating the pace and scale of cyberattacks targeting government networks, critical infrastructure, and businesses. The Australian Signals Directorate reported an 80% surge in security alerts over the past year, with over 4,400 warnings issued.

The agency attributes the rise in threats to AI tools enabling hackers to identify vulnerabilities more rapidly and carry out attacks with increased sophistication. State-sponsored hackers, particularly those sponsored by China, Russia, and North Korea, continue to target Australian organizations to gain access to sensitive or confidential information.

Additionally, the agency highlighted emerging risks from unintended actions of AI agents and model testing. Defence Minister Richard Marles emphasized the need for Australians to adapt to the rapid advancement of AI and strengthen their cybersecurity measures, urging individuals and organizations to update their systems.

Written by urgent.news from The Vibes's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.

Read the original at thevibes.com →

More in AI

Local-First AI Agent Testing: How to Expose Local MCP Servers Safely Without Compromising Host Security

Introduction: Local Tools, Remote Clients AI engineering has moved a lot of work onto developer machines. Agents call local tools, read local files, and query local databases through the Model Context…

  • Local-first AI agent testing enables model context protocol (MCP) agents on personal machines.
  • Exposing local MCP servers via tunneling poses security risks to host security.
  • Authenticated stdio servers use environment variables, not OAuth, for security.

More from Tuesday 27 October →