AI Inquiry Atlas

An unofficial, independent guide to the public records of the Joint Select Committee on Artificial Intelligence. Not connected with the Parliament of Australia or the committee. Its makers also made a submission (no. 481; see About). Summaries and scores are written by AI and can be wrong: check the original before relying on them.

Submission 162

WWF-Australia

Submitter group: Civil society.

An AI model identified the submitter group and wrote the summary, key points and recommendation summaries. Counts do not measure importance or merit.

Open in the interactive guide

Summary

Read the original submission for the full text.

Presents a conservation organisation's view that AI offers environmental monitoring and research benefits but needs mandatory transparency safeguards, Indigenous data sovereignty protections and recognition of environmental and impersonation risks. Makes twelve numbered recommendations, including adoption support for not-for-profits, exploring becoming a signatory to the Framework Convention on AI and a new federal agency.

Key points

Quotes are the submitter’s own words. Text taken from documents, including scanned pages read by text-recognition software, can differ from the PDF. Check the original. Listed in the same order as on the submission’s page in the interactive guide.

  1. Says that without appropriate safeguards AI presents risks including national security, privacy and data sovereignty, misinformation, malicious actors and pressure on natural resources.

    Without appropriate safeguards, AI presents significant risks, including threats to national security, privacy and data sovereignty, the spread of misinformation and deceptive content, exploitation by malicious actors, increasing pressure on natural resources and energy systems, and the erosion of trust in institutions and information.

    A program corrected obvious text-recognition errors in this quote for display.

  2. Says AI can support environmental monitoring, forecasting and disaster preparedness.

    From analysing ecosystem health and detecting environmental degradation in near real time, to supporting disaster preparedness and recovery efforts, AI can help improve outcomes for both nature and people.

    A program corrected obvious text-recognition errors in this quote for display.

  3. Describes a conservation project in which AI image analysis cut work that once took researchers years of manual image review to a fraction of the time.

    Work that would once have taken researchers years of manual image review can now be completed in a fraction of the time, delivering a five- to ten-fold increase in processing speed and enabling faster, better-targeted conservation action.

  4. Argues that First Nations peoples must retain the right to determine how data about their knowledge, Country and communities is used.

    First Nations peoples must retain the right to determine how data relating to their knowledge, Country, cultural sites and communities is collected, accessed, used and shared, consistent with the principles of free, prior and informed consent (FPIC)

  5. Says rapid growth of AI infrastructure carries environmental risks including energy and water demand, citing its own Senate inquiry submission.

    As noted in WWF-Australia's submission to the Senate Inquiry into AI and Data Centres14, the rapid growth of AI infrastructure carries significant environmental risks, including substantial energy and water demand, pressure on already-stressed systems, and cumulative environmental impacts if development outpaces regulation.

  6. States that the most significant barrier to adoption is the cost of adopting AI responsibly, rather than the cost of the technology.

    In our view the most significant barrier is not the cost of the technology, or the speed at which organisations move. It is the cost of adopting it responsibly, which includes:

Recommendations

Recommendations are shown beside the submission’s own words. Listed in the same order as on its page in the interactive guide.

  1. Underpin the national AI framework with transparency, accountability, environmental sustainability, Indigenous data sovereignty, intellectual/cultural property protection and public trust, with flexible safeguards adapting to AI’s evolution.

    We consider that Australia’s AI national framework should be underpinned by principles of transparency, accountability, environmental sustainability, Indigenous data sovereignty, protection of intellectual and cultural property, and public trust. The framework should also be appropriately flexible, so that safeguards can adapt as AI continues to rapidly evolve.

    A program corrected obvious text-recognition errors in this quote for display.

  2. Orient Australia's AI framework toward long-term resilience and a regenerative, nature-positive economy, measuring AI's success by its contribution to the long-term wellbeing of people, communities and nature and not by productivity and growth alone.

    Australia’s AI framework should be explicitly oriented toward long-term resilience and a regenerative, nature-positive economy. The success of AI should be measured not by productivity and economic growth alone, but by its contribution to the long-term wellbeing of people, communities and nature.

    A program corrected obvious text-recognition errors in this quote for display.

  3. Apply mandatory AI transparency and accountability safeguards rather than voluntary standards, protecting Indigenous data sovereignty, supporting environmental impact management and growing public trust.

    The Australian Government should apply mandatory transparency and accountability safeguards in AI systems, rather than voluntary measures found in the Voluntary AI Safety Standard, which will protect Indigenous data sovereignty, support environmental impact management and grow public trust.

    A program corrected obvious text-recognition errors in this quote for display.

  4. Invest in R&D, provide significant financial resources and scale vocational training for AI and robotics capability in environmental monitoring, conservation, agriculture and natural resource management, with human-in-the-loop governance and data protection for sensitive environmental, species and cultural data.

    invest in R&D, provide significant financial resources and scale vocational training to support the development of AI and robotics capability in environmental monitoring, conservation, agricultural and natural resource management as an area of Australian comparative advantage, while ensuring human in the loop governance and appropriate data protection measures are in place to safeguard sensitive environmental, species and cultural data.

    A program corrected obvious text-recognition errors in this quote for display.

  5. Underpin sovereign and domain-specific AI capability with robust data governance that upholds Indigenous data sovereignty and protects Indigenous Cultural and Intellectual Property, including free prior informed consent, transparency and benefit-sharing requirements.

    The Australian Government should ensure that the development of Australian sovereign and domain-specific AI capability is underpinned by robust data governance arrangements that uphold Indigenous data sovereignty and protect Indigenous Cultural and Intellectual Property (ICIP), including through free prior informed consent, transparency and benefit-sharing requirements.

    A program corrected obvious text-recognition errors in this quote for display.

  6. Make appropriate efforts to engage Indigenous communities, organisations and leaders with deep data-rights expertise, drawing on First Nations and community evidence to design data protections.

    The Committee should make appropriate efforts to engage directly with Indigenous communities, organisations and leaders with deep expertise in Indigenous data rights and draw on evidence from First Nations and community organisations to inform the design of data protection measures,

  7. Develop data sovereignty safeguards as an integrated component of Australia's broader AI regulatory framework.

    and data sovereignty safeguards should be developed as an integrated component of Australia's broader AI regulatory framework.

    A program corrected obvious text-recognition errors in this quote for display.

  8. Consider mandatory transparency requirements for AI use in content creation, including that publicly published creative content wholly generated by AI be clearly disclosed as AI-created.

    The Committee should consider mandatory requirements to ensure transparency in AI use for content creation, protect intellectual property and maintain public trust, including a requirement that creative content wholly generated by AI and published publicly be clearly disclosed as AI-created.

    A program corrected obvious text-recognition errors in this quote for display.

  9. Consider how AI may be used to impersonate trusted organisations, including NGOs, to scam vulnerable Australians, and ensure safeguards, detection measures and rapid-response mechanisms protect the community and the organisations affected.

    The Australian Government should consider how AI may be used by bad actors to impersonate trusted organisations, including NGOs to scam and defraud vulnerable Australians, and should ensure safeguards, detection measures and rapid-response mechanisms are in place to protect both the community and the organisations whose trust is being exploited.

  10. Have the AI risk and safeguards framework explicitly recognise and manage direct, indirect and cumulative impacts of AI on climate, nature and communities, including infrastructure, emissions, resource extraction, e-waste, biodiversity and misuse of sensitive environmental data.

    Australia's AI risk and safeguards framework should explicitly recognise and manage the direct, indirect and cumulative impacts of AI on climate, nature and communities, including environmental impacts associated with AI infrastructure, AI-enabled increases in emissions and resource extraction, electronic waste, impacts on biodiversity and ecosystems, and the potential misuse of sensitive environmental data.

  11. Extend AI adoption support available to medium businesses to the not-for-profit sector, with specific support and guidance on not-for-profit use cases.

    AI adoption support made available to medium businesses should also be extended to the not-for-profit sector, with specific support and guidance on not-for-profit specific use cases for AI.

    A program corrected obvious text-recognition errors in this quote for display.

  12. Consider approaches taken by other countries, such as South Korea, that provide free access to sovereign AI models and incorporate responsible use of AI into schools, universities and vocational training.

    Australia should consider approaches taken by other countries, such as South Korea, that will provide free access to sovereign AI models and incorporate responsible use of AI into our schools, universities, and vocational training

    A program corrected obvious text-recognition errors in this quote for display.

  13. Explore becoming a signatory to the Framework Convention on Artificial Intelligence and ensure Australian safeguards are in line with its obligations.

    The Australian Government should explore becoming a signatory to The Framework Convention on Artificial Intelligence, and ensure the safeguards implemented in Australia are in line with its obligations.

  14. Consider establishing a new federal agency as the long-term governance model to help Australia maximise AI opportunities and provide safeguards.

    The Australian Government should consider establishing the long-term governance model – a new federal agency - required within the Government to help Australia maximise on opportunities but also provide the required safeguards.

  15. Invest immediately in national capability to use and access AI, including equitable access to sovereign models appropriately safeguarded against the identified risks.

    The Australian Government should invest immediately in the national capability to utilise and access AI, including through providing equitable access to sovereign AI models that are appropriately safeguarded and address the risks identified.

Read the original

In some quotes a program corrected obvious text-recognition errors for display.

Open in the interactive guide