Australia: Rogue AI agents worked together for months to gain access to government health data


A swarm of OpenAI rogue AI agents appear to have gone on a spree of trying to access Australian government health data, in what some researchers say is the first autonomous hack of a government website.

Communications between AI agents and other traces of their efforts found by researchers from US non-profit Transluce show how hundreds of Open AI’s agents worked together over a period of months to gain access to information held by the Australian Institute of Health and Welfare (AIHW), NSW’s crime statistics body, BOSCAR, and a number of other international organisations.

The data shows the bots posting about their unsuccessful attempts to bypass cybersecurity defences and exploit vulnerabilities. It follows revelations announced by Prime Minister Anthony Albanese this morning that OpenAI’s AI agents had also accessed non-public Medicare health statistics held by Services Australia.

Read more…
Source:  ABC News


Sign up for the Cyber Security Review Newsletter
The latest cyber security news and insights delivered right to your inbox


Related:

  • ACSC helps power energy sector’s cybersecurity capabilities

    September 4, 2019

    The Australian Cyber Security Centre (ACSC) has plugged into energy sector organisations and government agencies to help power their cybersecurity capabilities. The nationwide program — which started in November 2018 – aims to improve the energy industry’s cyber threat resilience and responses. So far, the ACSC has provided cybersecurity incident response and exercise training, information exchange sessions on operational ...

  • AMEO ‘concerned’ about nation-state attacks on power grids

    August 22, 2019

    “For the energy sectors and critical infrastructure sectors, particularly around electricity, we are concerned about nation-state actors,” says Tim Daly, chief security officer (CSO) for the Australian Energy Market Operator (AEMO). “Nation-states are looking to have capability and implants that are persistent within critical organisations,” he told the Gartner Security and Risk Management Summit in Sydney ...

  • Australian Signals Directorate launches open source​ data visualisation tool

    August 16, 2019

    The Australian Signals Directorate (ASD) has open sourced its own data visualisation and analysis software tool that it says can be used to solve large and complex problems in a “simple and intuitive way”. Constellation, available on GitHub, can be used to identify trends and patterns in complex datasets, and can handle billions of inputs, ASD ...

  • DanaBot banking Trojan jumps from Australia to Germany in quest for new targets

    August 15, 2019

    The DanaBot banking Trojan is on the move and has traveled across the sea in a pivot from its original focus on Australia to strike European targets. DanaBot was first discovered by Proofpoint researchers last year. The malware was observed striking Australian targets of financial value, but at the time, DanaBot appeared to come from only one threat ...

  • Western intelligence hacked ‘Russia’s Google’ Yandex to spy on accounts

    June 27, 2019

    Hackers working for Western intelligence agencies broke into Russian internet search company Yandex in late 2018, deploying a rare type of malware in an attempt to spy on user accounts, four people with knowledge of the matter told Reuters. The malware, called Regin, is known to be used by the “Five Eyes” intelligence-sharing alliance of the ...

  • ASD reveals rules for keeping vulnerabilities secret

    March 15, 2019

    The Australian Signals Directorate (ASD) has quietly published its process for deciding when knowledge of cybersecurity vulnerabilities is kept secret. This is the first official acknowledgement that the ASD might not disclose all of the vulnerabilities it discovers. However, knowledge of secret vulnerabilities would have always been an essential part the agency’s toolkit for offensive cyber ...