ASD Threat Report Says AI Drove an 80 Per Cent Jump in Cyber Warnings as Disruption Becomes Inevitable

The Australian Signals Directorate has warned Australians to prepare for more frequent disruption to everyday services as artificial intelligence turbocharges cyber attacks, ABC News reported on October 27, as the agency released its annual cyber threat report. Director-General Abigail Bradshaw said frontier AI models had “redefined” the threat facing the country. “What is inevitable is that with each capability increase, we are likely to see more disruption,” she told the ABC, pointing to websites that had to be taken offline in recent weeks so they could be patched.

The report calls AI “the defining force in the cyber threat landscape” over the past year, allowing malicious actors to find vulnerabilities faster and exploit them with unprecedented scale, speed and sophistication, SBS News reported, citing AFP. The directorate issued more than 4,400 warnings about potentially malicious activity, an 80 per cent increase on the previous 12 months. State-sponsored actors kept targeting Australian organisations, critical infrastructure, businesses and individuals for sensitive information, and hackers linked to China, Russia and North Korea targeted Australia or its allies. The report also notes new threats from the unintended actions of AI agents and from model testing.

The ASD said it had used some AI tools to compile the report itself, and the ABC said the agency also disclosed previously secret details of offensive operations, including the targeting of criminals who hoard and sell personal data on the dark web. Defence Minister Richard Marles told reporters there was no turning back the clock. “AI is here, and it’s real,” he said, urging Australians to keep their computers updated and harden their systems.

What it means in practice

ASD Threat Report Says AI Drove an 80 Per Cent Jump in Cyber Warnings as Disruption Becomes Inevitable — contextual photo

For organisations, the message is that patch windows are shrinking. The recent scramble over AI-found ‘PHP2shell’ flaws that pushed GovCMS and Moodle offline is the kind of disruption Bradshaw describes, and it follows her earlier call for an AI early-warning system for Australia. Boards should expect more emergency patching, more short outages taken deliberately to close holes, and sharper questions about which AI agents can reach internal systems. Federal policy is already moving that way after Home Affairs ordered AI cyber reviews following the Medicare breach.

What to watch next is whether the government ties the report’s findings to new funding or mandatory rules, and how quickly critical infrastructure operators can shorten the gap between a disclosed flaw and a deployed fix.