Bureau of Meteorology

# Artificial Intelligence (AI) Transparency Statement ## Overview This statement accords with the Digital Transformation Agency’'s (DTA) Policy for responsible use of AI in government. ## Definition The Bureau applies the Organisation for Economic Co-operation and Development (OECD) definition: An AI system is a machine-based system that, for explicit or implicit objectives, infers, from the input it receives, how to generate outputs such as predictions, content, recommendations, or decisions that can influence physical or virtual environments. Different AI systems vary in their levels of autonomy and adaptiveness after deployment. ## Recognised benefits The Bureau has benefited from limited use of AI. For example, we took part in the DTA's 2024 whole-of-government trial of Copilot for Microsoft 365. We acknowledge the potential benefits of further research and adoption of AI technologies. We commit to: • - safe and responsible use of AI •- ensuring AI implementation aligns with Australian Public Service values and prioritises staff wellbeing. ## How we use AI The Bureau applies the Australian Governments classification system to categorise: • - hhow ow AI is used, and •- tthe he domains in which it is applied. ### Domains We may use AI across these domains: • - scientific\ •- corporate and enabling ### Use We may use AI in the following ways.\ #### Automate or simplify routine tasks\ • - Using AI to help our organisation to be more productive, by: o - supporting workflow management,\ o - reducing administrative burden, and\ o - enhancing communications. #### Support decision making and administrative actions\ • - Using AI to guide, assess or make recommendations to a human decision maker.\ #### Identify, understand and present insights • - Using AI to support data analysis and reporting. #### Support software development • - Using AI to assist technical staff with drafting and reviewing code, while maintaining human oversight and accountability. ## AI in forecasting The Bureau has used computer science to prepare data for forecasts since the 1970s. Computer science is broader than the field of AI.\ We use advanced computational science, such as advanced statistical techniques and computational physics, to: • - process large volumes of data, and •- perform complex calculations to produce forecast outputs. We also use machine learning to calibrate, verify and automate these outputs. Our scientists and meteorologists provide quality control as appropriate. Before going into production, all computer science practices are subject to rigorous: • - research, •- testing, and •- assurance.\ This includes machine learning and AI in forecasting services. All key applications used to generate forecasts are built specifically for our use. They are built, maintained and managed in-house. We do not use neural network systems – 'generative AI' – for weather forecasting outputs. The suitability of these techniques is an active field of research at the Bureau. ## Public interaction and impact We commit to ensure AI does not negatively impact our customers, including the public. We will adopt a gradual approach to any future use of AI modelling methods. Our intent is to improve the impact and value of our services. Humans review, validate and authorise any public content created or supported by AI. This aims to ensure it is fair and accountable.\ ## Monitoring\ Our executive team oversees the Bureau's strategic direction and use of AI. This involves regular review to ensure responsible and effective use.\ ## Compliance The Bureau has adopted Australia’'s 8 AI ethics principles. We have internal guidelines to ensure AI is used in line with these principles.\ The guidelines include activities to help the Bureau apply each principle in practice. These activities are based on the Department of Finance's National Assurance Framework.\ The Bureau will: • - use AI according to relevant legislation, frameworks and policies •- follow mandatory requirements in the Policy for responsible use of AI in government. We will regularly review compliance to the Australian Government's AI policy landscape. This will happen through existing governance and environmental scanning practices.\ AI use in the Bureau's forecasting and prediction services: • - accords with research guidelines, and\ •- complies with the usual standard of scientific scrutiny.\ ## Accountable Official The Bureau's Chief Information and Technology Officer is the AI Accountable Official. This appointment was made in November 2024.\ ## Contact for AI enquiries For questions about this statement, contact us. ## Updates to this statement This statement was last updated on 2 March 2026. We review and update this statement yearly, or when: • - our approach to AI changes, or •- there is material change to its accuracy.

Artificial Intelligence (AI) Transparency Statement Overview This statement accords with the Digital Transformation Agency’s (DTA) Policy for responsible use of AI in government. Definition The Bureau applies the Organisation for Economic Co-operation and Development (OECD) definition: An AI system is a machine-based system that, for explicit or implicit objectives, infers, from the input it receives, how to generate outputs such as predictions, content, recommendations, or decisions that can influence physical or virtual environments. Different AI systems vary in their levels of autonomy and adaptiveness after deployment. Recognised benefits The Bureau has benefited from limited use of AI. For example, we took part in the DTA's 2024 whole-of-government trial of Copilot for Microsoft 365. We acknowledge the potential benefits of further research and adoption of AI technologies. We commit to: • safe and responsible use of AI • ensuring AI implementation aligns with Australian Public Service values and prioritises staff wellbeing. How we use AI The Bureau applies the Australian Governments classification system to categorise: • h ow AI is used, and • t he domains in which it is applied. Domains We may use AI across these domains: • scientific \ • corporate and enabling Use We may use AI in the following ways. \ Automate or simplify routine tasks \ • Using AI to help our organisation to be more productive, by: o supporting workflow management, \ o reducing administrative burden, and \ o enhancing communications. Support decision making and administrative actions \ • Using AI to guide, assess or make recommendations to a human decision maker. \ Identify, understand and present insights • Using AI to support data analysis and reporting. Support software development • Using AI to assist technical staff with drafting and reviewing code, while maintaining human oversight and accountability. AI in forecasting The Bureau has used computer science to prepare data for forecasts since the 1970s. Computer science is broader than the field of AI. \ We use advanced computational science, such as advanced statistical techniques and computational physics, to: • process large volumes of data, and • perform complex calculations to produce forecast outputs. We also use machine learning to calibrate, verify and automate these outputs. Our scientists and meteorologists provide quality control as appropriate. Before going into production, all computer science practices are subject to rigorous: • research, • testing, and • assurance. \ This includes machine learning and AI in forecasting services. All key applications used to generate forecasts are built specifically for our use. They are built, maintained and managed in-house. We do not use neural network systems – 'generative AI' – for weather forecasting outputs. The suitability of these techniques is an active field of research at the Bureau. Public interaction and impact We commit to ensure AI does not negatively impact our customers, including the public. We will adopt a gradual approach to any future use of AI modelling methods. Our intent is to improve the impact and value of our services. Humans review, validate and authorise any public content created or supported by AI. This aims to ensure it is fair and accountable. \ Monitoring \ Our executive team oversees the Bureau's strategic direction and use of AI. This involves regular review to ensure responsible and effective use. \ Compliance The Bureau has adopted Australia’s 8 AI ethics principles. We have internal guidelines to ensure AI is used in line with these principles. \ The guidelines include activities to help the Bureau apply each principle in practice. These activities are based on the Department of Finance's National Assurance Framework. \ The Bureau will: • use AI according to relevant legislation, frameworks and policies • follow mandatory requirements in the Policy for responsible use of AI in government. We will regularly review compliance to the Australian Government's AI policy landscape. This will happen through existing governance and environmental scanning practices. \ AI use in the Bureau's forecasting and prediction services: • accords with research guidelines, and \ • complies with the usual standard of scientific scrutiny. \ Accountable Official The Bureau's Chief Information and Technology Officer is the AI Accountable Official. This appointment was made in November 2024. \ Contact for AI enquiries For questions about this statement, contact us. Updates to this statement This statement was last updated on 2 March 2026. We review and update this statement yearly, or when: • our approach to AI changes, or • there is material change to its accuracy.