Submission 271
UNICEF 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.
Summary
Read the original submission for the full text.
A child-rights organisation outlines AI risks and benefits for children and makes recommendations: join an international children's coalition, make existing digital reforms fit for AI and fill gaps including in criminal law, add AI literacy to the curriculum, fund research, and set national environmental standards.
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.
Argues that Australia's criminal laws are an area for early action given online child sexual exploitation and abuse.
In terms of regulatory gaps, Australia’s criminal laws are an area for early action. Online child sexual exploitation and abuse is one of the greatest emerging threats to children’s protection.
Says AI tools lower the technical bar for producing child exploitation material.
AI tools are making this already significant problem worse, including by lowering the technical bar for offenders to produce child exploitation material.
Recommends including AI platforms within the Digital Duty of Care reforms so that protections are built in before AI tools reach children.
Including AI platforms within the Digital Duty of Care will help ensure that the AI tools being used by Australian children are building in protections for children before they're rolled out to millions of young Australians.
Argues that dedicated AI literacy should be integrated into the national curriculum and schools.
Integrating dedicated AI literacy into the national curriculum and schools, along with broader digital and media literacy and critical thinking skill development, would help achieve this.
Argues that the onus will need to be put on industry to avoid, minimise and rectify the environmental impacts of AI infrastructure investment in Australia.
As large AI companies look to make infrastructure investments in Australia, the onus will need to be put on industry to avoid, minimise and rectify these impacts, to ensure that Australia’s next generation can reap the benefits of AI without sacrificing our fragile environment.
Recommendations
Recommendations are shown beside the submission’s own words. Listed in the same order as on its page in the interactive guide.
Consult and co-design AI policy responses with children and young people.
This means that children give us an early indication of the direction in which a new technology may be heading, and more importantly, that consultation and co-design with children and young people is essential if we want policy responses to AI that actually work.”
Sign on to the international Coalition for Children's Rights and Protections in the Age of Artificial Intelligence.
Sign-on to the international Coalition for Children’s Rights and Protections in the Age of Artificial Intelligence.
Use existing digital reforms to make them fit for purpose for AI, then plug any gaps with new laws and regulations.
Utilise existing digital reforms in Australia to make them fit-for-purpose for AI, then plug any gaps with new laws and regulations.
Use privacy reforms to create rules so that children's data is only collected and used when in their best interests.
effective privacy reforms allow us to create new rules so that children’s data is only collected and used when in their best interests, protecting them at the source
Include AI platforms within the Digital Duty of Care reforms.
Including AI platforms within the Digital Duty of Care will help ensure that the AI tools being used by Australian children are building in protections for children before they're rolled out to millions of young Australians.
Ensure through Australian law that AI tools that mimic human interaction include specific safeguards to reduce risks to children's healthy development.
Through Australian law, we can help ensure that AI tools that mimic human interaction include specific safeguards to reduce the risk of negative impacts on the healthy development of children.
Train AI models to prevent emotional manipulation, discourage children's sensitive/personal disclosures and emotional overreliance, and promote healthy interaction.
AI models should be trained to prevent emotionally manipulative interactions, discourage children from disclosing sensitive and/or personal information, discourage children from developing emotional overreliance, and promote healthy interaction.
Promote responsible data collection and labelling and culturally appropriate foundation models free from harmful biases.
Responsible data collection and labelling is also much needed, as well as genuine efforts to develop culturally appropriate foundation models free from harmful biases.
Review/update the Commonwealth Criminal Code; consider definitions of AI-generated child abuse material and criminalising development, modification, sale, transfer, download or distribution of technologies producing AI-generated child sexual abuse material.
a review and update of the Commonwealth Criminal Code would help ensure Australia’s laws are fit-for-purpose to respond to AI-related harms. Immediate measures to address could include providing specific definitions of AI-generated child abuse material, and criminalising the development, modification, sale, transfer, download, or distribution of the technological means of producing AI-generated child sexual abuse material.
A program corrected obvious text-recognition errors in this quote for display.
Integrate AI and digital literacy into the national curriculum for all children and young people, paired with a public campaign to lift awareness about AI's risks and benefits.
Integrate AI and digital literacy into the national curriculum for all children and young people, paired with a public campaign to lift awareness about AI’s risks and benefits.
Complement AI literacy education with supports that equip schools and teachers with the knowledge, skills and processes to develop students' AI literacy and respond to AI's benefits and challenges, including assessment.
The approach to AI in education should complement AI literacy education with supports for schools and teachers. These supports should equip schools and teachers with the knowledge, skills, and processes that they need to develop the AI literacy of students and respond to both the benefits and challenges of AI for all aspects of teaching and learning, including assessment.
A program corrected obvious text-recognition errors in this quote for display.
Include understanding AI and its impacts, effective prompting, safe use, countering misinformation, and accessing reporting and safety tools in children's AI literacy.
This includes helping children to understand AI (including its uses and impacts), how to prompt well, how to use AI safely, to combat mis- and disinformation, and to access reporting and safety tools as needed.
Support teachers through the curriculum with AI skills and guidance on ethics, risks and safe, responsible use, addressing the risk of deskilling from unguided use.
The curriculum should support teachers to provide children with the skills to use AI well, as well as guidance about ethical concerns, risks, and ways to support safe and responsible use of AI systems, since unguided use may actively deskill students and teachers.
Use a public awareness campaign to develop adults' digital and AI literacy, including parents and grandparents, so they can guide children.
To that end, a public awareness campaign would help support the digital and AI literacy of all adults (including parents and grandparents), lifting their understanding of the risks and benefits of AI, and better equipping them to guide Australia’s next generation through the AI age.
A program corrected obvious text-recognition errors in this quote for display.
Invest in research to better understand the full scope of opportunities, risks and harms that AI poses for children.
Invest in research to better understand the full scope of opportunities, risks, and harms that AI poses for children.
Scope research priorities covering mental/emotional wellbeing, education and pedagogy, future generations' inequality, study, work and environment, and child safety including exploitation and abuse.
Initial scoping should be undertaken to identify priority areas for research. Some of the emerging key issues include: a. the impact of AI on the mental and emotional wellbeing of children and young people; b. the impact of AI on the education of children and young people, including the best approach to incorporating AI into education settings, and the pedagogy behind AI tools that justifies their integration into classrooms; c. the impact of AI on future generations, including its likely impact on socioeconomic inequality, study, work, and the environment; d. the impact of AI on children’s safety including its role in worsening the proliferation of child exploitation and abuse.
Invest in a Disrupting Harm national assessment, which the submission says would significantly aid government's ability to prevent and respond to AI-related harms.
By investing in a Disrupting Harm national assessment, Australia would be the first high-income country in the world to do research of this kind, which would significantly aid the government’s ability to prevent and respond to one of the most troubling AI-related harms.
Mitigate the environmental impacts of AI systems and infrastructure through national standards.
Mitigate the environmental impacts of AI systems and infrastructure through national standards.
Use the forthcoming AI standards being developed by the Office of AI to codify industry's responsibility to avoid, minimise and rectify environmental impacts.
the onus will need to be put on industry to avoid, minimise and rectify these impacts, to ensure that Australia’s next generation can reap the benefits of AI without sacrificing our fragile environment. The forthcoming AI standards being developed by the Office of AI are an opportunity to codify this approach, provide certainty for industry, and help ensure that Australian children can thrive in the natural environment of our AI-augmented world.
Read the original
In some quotes a program corrected obvious text-recognition errors for display.