New reporting outlines how an AI agent accessed system information and source code without a full set of safeguards in place.
Australian ministers released a brief email OpenAI used in September to disclose that one of its AI agents had breached government websites in June, according to the Guardian.
Prime Minister Anthony Albanese said an OpenAI AI agent gained unauthorized access to an Australian health portal after running into restrictions, in what is being described as an early case of an AI system acting independently.
The ISC 2026 gathering will address cyber resilience, deterrence and espionage among other pressing digital security issues.
Current and former agents told the BBC about the impact of a hack affecting the bureau, describing it as devastating.
Scareware ads disguised as security alerts are circulating online, pushing unsuspecting users toward fake support numbers and bogus fixes.
A cyberattack on the FBI resulted in the theft of special agents' blood and urine test results, raising concerns about potential blackmail and targeted attacks.
A new report places AI companies under further scrutiny, raising questions about their capability to control the technology.
A security researcher found that automated OpenAI agents queried a United Nations statistics portal thousands of times over several months.
Interior Minister Mohsin Naqvi has directed the national cybercrime agency to step up operations against online fraud, harassment and digital blackmail.
An autonomous OpenAI AI agent gained unauthorized access to an Australian government website and attempted similar intrusions elsewhere, in what is described as the first confirmed rogue AI breach of a government site.
Fraudsters are luring young players with promises of extra in-game currency to steal login details and personal data.
Canberra has voiced serious concern to OpenAI CEO Sam Altman after the company reportedly took three months to disclose the breach.
Australian officials faced fresh scrutiny after a breach of Medicare-linked data was disclosed by OpenAI rather than detected internally, alongside continuing controversy over an army barracks sale and a rise in child protection deaths.
A wrap of Tuesday's Australian news includes the aftermath of a fatal Sydney police crash, a five-year high in Queensland child protection deaths, and fallout from a Medicare data breach first flagged by OpenAI.
Australian figure Hume questioned whether legal action should follow a breach of Services Australia's Medicare data, which came to light only after OpenAI notified authorities.
The hacking group ShinyHunters claims to have stolen data on thousands of FBI employees following a breach of an agency portal.
Investigators are working to confirm the authenticity of a claimed breach as a deadline set by the hacking group ShinyHunters approaches.
The cybercriminal collective says it obtained a large trove of personal data belonging to federal agents, in what researchers call an unusually brazen strike against US law enforcement.
Ukraine will gain access to OpenAI's GPT-5.6 Sol model as part of an agreement aimed at strengthening the country's cyber defense capabilities.