Submission 337
Name withheld
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.
It is clear that these huge electrical demands can only delay Australia’s progress towards carbon-free electricity.
It is clear that these ‘AI’ datacentres are something new. At least ten times more power footprint than previous types of datacentres. Based on the number of backup diesel generators (a typical datacentre of past years would have one or two) it looks like these ‘AI’ datacentres use around fifty times more power than previous datacentres.
the scrapers use ‘botnets’ of criminally subverted computers, home routers, and downloadable applications such as VPNs or web browsers. Those applications may themsleves be ordinary applications which have been criminally subverted: ‘trojan horse software’. In this way even highly regarded ‘AI’ firms have links to organised crime at one remove.
‘Fair dealing’ does not include the training for large language models or the operation of the resulting large language model by the public.
The parliament should think carefully before extending fair dealing to AI firms. It would be outrageous that international for-profit companies could access copyrighted content for free which Australia’s disabled and disadvantaged people have to pay for.
Some commentators claim that LLMs and datacentres have the attributes of a financial bubble. Ed Zitron is a proponent of this view and I would suggest the committee read one of his summary articles or speeches. I am a professional economist and are in general agreement with his view.
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.
The Australian government should take measures to prevent the use of criminally-obtained computer-time by LLM firms which also acquire datacentre space in Australia.
The parliament should think carefully before extending fair dealing to AI firms. It would be outrageous that international for-profit companies could access copyrighted content for free which Australia’s disabled and disadvantaged people have to pay for.
Some commentators claim that LLMs and datacentres have the attributes of a financial bubble. Ed Zitron is a proponent of this view and I would suggest the committee read one of his summary articles or speeches.