Submission 471
Mr Michael Huang
Submitter group: Individuals.
Position labels are not shown for individual submitters.
An AI model identified the submitter group and chose the extracts. Counts do not measure importance or merit.
Extracts from the submission, in the submitter’s own words.
Extracts
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.
This statement was signed by Geoffrey Hinton and Yoshua Bengio, two Turing Award winners known for their pioneering work in the field. Bengio estimated a “20 per cent probability that it [AI] turns out catastrophic.” Hinton’s estimate is that “the risk is more than 50%”.
Loss of control scenarios are scenarios in which one or more general-purpose AI systems operate outside of anyone’s control, and regaining control is either extremely costly or impossible.
OpenAI’s investigation into itself was limited in staffing, time and scope. Elizabeth Barnes, CEO of METR, reported that only 3 independent staff had access to OpenAI offices and documents, and the investigation was limited to only 6 days.
American, British and Australian legislators also want to make AI kill switches mandatory (e.g. U.S. AI Kill Switch Act, U.K. Amendment to the Cyber Security and Resilience Bill, Australian AI Kill Switch and Data Centre Control Bill 2026).
Extracts from recommendations
Extracts chosen by an AI model, in the submitter’s own words. Listed in the same order as on its page in the interactive guide.
Establish a federal royal commission into AI
Bring representatives of the U.S. and China together to agree to pause frontier AI development, include mandatory kill switches, and establish moratoriums on automated recursive self-improvement and artificial super intelligence. Formalise the agreement through a legally binding, verifiable international treaty and international agency.
Pass Australian federal legislation to pause frontier AI development, make kill switches mandatory, and establish moratoriums on automated recursive self-improvement and artificial super intelligence. Create a federal AI regulator (or provide regulatory powers to the Australian AI Safety Institute) through legislation.