Submission 395
Ms Grace Adams
Submitter group: Individuals.
Position labels are not shown for individual submitters.
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.
Writing as a charity CEO in a personal capacity, expresses concern about AI-enabled cyber attacks, job losses and global catastrophic risks. Asks the government to consider amending copyright law for onshore model training, mandatory incident reporting for LLMs, and international cooperation to slow or pause AI development.
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.
Says they are concerned that their charity could not withstand future cyber attacks, risking donor data and funds.
I am concerned that our charity will not be able to withstand an influx of future cyber attacks, and may lead to the stealing of our donor’s data, or worse, stealing of funds that were intended to improve the lives of people in extreme poverty.
Says they are concerned about AI's potential effect on jobs and societal cohesion, and about bioweapons and concentration of authoritarian power.
I am also concerned about AI’s potential impact on job losses and societal cohesion, as well as more global risks, such as bad actors developing bioweapons, or increasing the risk of concentration of authoritarian powers globally.
Argues that government needs to take strong action to regulate AI risks while capabilities are still being understood.
I believe that the government needs to take strong action to regulate the risks of AI, while we are still understanding the capabilities and potential ramifications of this technology that is improving at an exponential rate.
Argues that making model training possible and attractive onshore would support strategic regulatory oversight of capabilities, risks and incidents.
Finding a pathway to make training of models possible, and attractive onshore in Australia, would allow the government to strategically regulate important oversight over the capabilities, risks, and importantly, any incidents that these models produce, so that we may be able to take stronger action in decisive moments.
Recommendations
Recommendations are shown beside the submission’s own words. Listed in the same order as on its page in the interactive guide.
Consider copyright-law amendments allowing onshore AI-model training on the basis that this enables further Australian oversight and regulation of development.
I am asking that the government consider: - Pathways to amend copyright law to allow for onshore training of AI models, on the basis that this would allow Australia to further regulate the oversight and development of this technology
A program corrected obvious text-recognition errors in this quote for display.
Consider mandatory incident reporting for all LLMs and associated technologies.
Mandatory incident reporting for all LLMs, and associated technologies
Consider urgent cooperation with governments and AI companies globally to slow or pause AI development.
Urgent cooperation with key actors, including governments, and the AI companies themselves across the globe to slow down or pause development of AI
Read the original
In some quotes a program corrected obvious text-recognition errors for display.