Australia's Spy Chief Orders Security Over Uptime as Cyber Incidents Jump

The Australian Signals Directorate released its annual cyber threat report Tuesday, warning that keeping services online must yield to locking down system security. "System security must be prioritised over system availability," the agency said, adding that more online services will go offline unexpectedly as they are patched.

The directive arrives as the agency's own numbers climbed. In the 2025-26 year, ASD responded to 1,500 cybersecurity incidents, up 20% from the prior year. It also flagged more than 4,400 potential malicious incidents at outside organizations including businesses, governments and critical infrastructure, a jump of 80% in twelve months. "In a threat environment transformed by AI, collective action is our strongest defence," director general Abigail Bradshaw said.

ASD attributed the rise to hostile foreign governments, criminal groups and hackers using AI to gather intelligence, disrupt critical infrastructure and steal money. The agency said government and private networks in Australia faced persistent breach attempts from state-based actors last year, and said such actors "in the Indo-Pacific" could be preparing for future conflicts, including "future intelligence collection or disruption activities". The report stated that all network owners face elevated threat levels and that cybersecurity is a core business risk and leadership responsibility for every organization.

Bradshaw and Defence Minister Richard Marles told organizations to put security ahead of uptime, which could mean more outages while systems are secured. "We need AI tools to help automate our defences to stay ahead of AI-enabled attackers," Bradshaw said Monday. "Australia must accelerate the modernisation of its technology and cybersecurity foundations to protect the systems we rely on every day." Marles said Australians must enhance cyber defences while harnessing AI's benefits.

The agency said AI can find weaknesses or suspicious behavior and improve responses, and it called on businesses and critical infrastructure to deploy advanced AI models for defence. Bradshaw has said ASD has access to Anthropic's Mythos model, a tool the lab has not released publicly because it can identify unknown flaws in IT systems. Australian government agencies consider access to such frontier models and cooperation with leading AI labs critical to cybersecurity.

Bradshaw identified legacy technology, including outdated systems or devices no longer supported with updates by their makers, as among the largest risks and the most attractive targets for hackers. She said ASD wants organisations to modernise their technology and cybersecurity capabilities, and take stronger action against new threats. Home Affairs and Cybersecurity Minister Tony Burke said the government is positioning Australia to respond to threats and opportunities from emerging technologies.

ASD urged organisations to update systems, seek help from the agency and report suspicious incidents. Last week, newly discovered vulnerabilities in PHP, the programming language that underpins most websites, triggered a rush by governments and businesses to patch the flaw.

The annual threat report was released on Tuesday.