The Australian Signals Directorate’s director-general, Abigail Bradshaw, has commented on the implications of the ASD’s seventh Annual Cyber Threat Report, released today, warning that advanced AI models are redefining the threat landscape.
"What is inevitable is that with each capability increase, we are likely to see more disruption," Bradshaw (pictured) said in a widely reported statement.
"Disruption can look like what we've seen over the last few weeks, with websites that require patching having to be taken down in order to be secured.”
According to Bradshaw, these disruptions could take many forms, from data breaches to disruptions of vital services.
"Virtually any service that you rely on in your life today is enabled in some respect by a digital network."
The report revealed an alarming increase in the number of incidents fielded by the Australian Cyber Security Centre, and an 80 per cent increase in the number of incident notifications shared by the agency.
Tom Mooney, Director of the Democracy Counts alliance, said that while Australia’s cyber agencies are doing good work, the government must put the onus on tech companies themselves to do right by the Australian people in the first place.
"Artificial intelligence is not ungovernable. Australia has a sovereign right to decide how these technologies are used here, and citizens have a democratic right for our leaders to shape the AI transition in a way that puts the public interest before all else," Mooney said in a statement.
He noted recent AI-powered attacks, such as an OpenAI model accessing the Medicare system, and an incident that saw supermarket fridges in France overheat and spoil food, causing a public health threat and even starting fires in some locations.
“Our agencies report that they are working constructively with the companies developing frontier AI, and this presents a real opportunity for Australia to lead," Mooney said.
"The Australian Government needs to ensure companies that develop and deploy AI are accountable to Australians and subject to the law, and we have the opportunity to build a coalition of like-minded countries and tech companies to support establishment of such a regime.”
Andrew Kay, Senior Director of Systems Engineering, APJ at Illumio, said that Australian organisations should treat Bradshaw's warning as a "planning assumption".
"Attackers will get in. AI just means they get in faster, using the same old gaps: legacy systems, missed patches, reused credentials," he said.
"What decides whether an intrusion becomes a headline is how far the attacker can travel once they’re inside, and that’s an architectural decision you make well before any attack."
Want to see more stories from trusted news sources?Make Cyber Daily a preferred news source on Google.
