AI transparency statement

We are dedicated to the responsible, safe, and transparent use of Artificial Intelligence (AI). The ŮŮÊÓƵAI Transparency Statement strives to ensure continuous transparency in our use of AI as we explore opportunities and implement AI capabilities. This commitment aligns with the Digital Transformation Agency's (DTA) .

How we use AI

ŮŮÊÓƵhas long been engaged in the exploration of innovative AI techniques to further support the delivery of scientific information. ŮŮÊÓƵhas applied Machine Learning to process, analyse and synthesise scientific data, particularly in Satellite Imagery and Geological Information, for many years. We are in the early stages of exploring broader artificial intelligence technologies to provide an enhanced level of efficiencies and capabilities, within our work practices and scientific disciplines, focussing on internal testing and exploration with no use cases being used on published products and services.

Use patterns and domains

ŮŮÊÓƵis investigating and employing AI systems for Scientific, Corporate and Enabling domains, and as a subset within each, the Service Delivery domain.

Scientific Research and Information
Usage Patterns: Analytics for insights, Image processing

The investigation and application of AI involve multiple scientific disciplines, including Critical Minerals, Satellite Imagery, Uncover-ML, Passive Seismic, Marine, Earthquakes, etc. The investigation of Generative AI technologies also aims to enhance user experience and improve access to Geoscience Australia’s scientific resources.

Examples of our utilisation of AI include tailored Machine Learning products made available to both internal and external stakeholders. Additionally, in support of Critical Minerals initiatives, current investigations are focused on building covariate libraries, large dataset analyses, and geochemical and geophysical modelling, all leveraging the High-Performance Computing capabilities at the National Computational Infrastructure (NCI).

Work Process Improvement AI
Usage Patterns: Decision making and administrative action, Workplace productivity

We are enhancing workplace efficiency with the use of Generative AI (such as Microsoft CoPilot) by automating administration tasks, accessing information quickly through improved user experience, letting staff focus on core tasks.

Compliance

We will utilise AI services in accordance with applicable legislation, regulations, frameworks and policies.

Accountable officials

In alignment with DTA’s , we have two accountable officials under the policy. These are the Chief Information Officer (CIO) and the Chief Scientist represented by those within each role.

Authoritative source, individual responsibility and monitoring

Quality assurance promotes transparency and accountability in AI usage.

ŮŮÊÓƵhighlights the importance of AI training, with a focus on accountability. As a government entity, it serves as an authoritative source when presenting critical information. Officials are required to ensure rigorous quality assurance checks to maintain the integrity and reliability of all systems, including AI.

Generative AI can be useful for research, summarising information, or creating content from known sources, but it should not generate messages or summarise unverified information without consent. AI-generated material must undergo thorough quality checks and should not be treated as the final source of truth; instead, it should be verified and cited by the original source.

The use of AI within ŮŮÊÓƵis monitored through an AI register and assessed (when applicable) through multiple processes with a focus on business cases, security and efficiency.

Reviews and more information

The transparency statement will be reviewed at least every twelve months, or as our approach to AI changes.

For more information or enquires about Geoscience Australia’s adoption of AI, you can contact us.