Opinion
Tech lords are promising us utopia. Their brave new world might be a dump
Melanie La'Brooy
WriterA number of recent humiliating fiascos have reinforced artificial intelligence’s growing image as the 21st century reincarnation of Tulip Mania.
In July, Elon Musk’s chatbot, Grok, was updated and promptly started spewing antisemitic and other toxic content.
Billionaire Scott Farquhar is arguing that writers, who earn an average $18,000 a year, should allow AI companies to use their work for free.Credit: Dominic Lorrimer
Earlier this month was the disastrous launch of OpenAI’s ChatGPT-5. OpenAI CEO Sam Altman had promised users that it would be like talking to “a PhD level expert in anything”, but within hours of the launch, epic fails started to flood in.
One user asked GPT-5 to generate a map of the United States with each state named, which is how we all learned about the great states of Aphadris, Wiscubsjia and Misfrani. It also had problems counting to 12, and referred to President Gearge Washingion. These kinds of inaccuracies are initially hilarious, until we realise we’re drowning in a sea of online misinformation and the joke’s on us.
Nevertheless, we’re repeatedly told that the AI spaceship is leaving for a brave new world, so we’d better get on board or risk being left behind. Unfortunately, the people steering the spaceship appear to have lost their moral compass. So where exactly we’re headed remains unclear.
Australia does not currently have AI-specific legislation. Chair of the Tech Council of Australia, Scott Farquhar, prefers it that way, stating that he doesn’t want Australia to be “hampered by the wrong legislation”.
The right legislation, according to the council, is a text and data mining exemption to the Copyright Act, which would allow AI companies to use copyrighted work to train their large language models without seeking consent or paying authors a cent. Their illogical argument is that the work of Australian artists is immeasurably valuable, while simultaneously worth nothing at all.
The council’s lobbying effort at the Economic Reform Roundtable this week will also push for Australia to build more data centres, the huge energy- and water-guzzling facilities which provide the vast power, storage and cooling requirements that AI requires. Farquhar has repeatedly argued that Australia should become a regional data centre hub, saying: “I think we are going to have a huge amount of benefits (from AI) and I hope we as a nation set ourselves up to have some of those benefits accrue to Australia.”
At least two Australians are likely to benefit from Australian AI data centres. Skip Capital, the company Farquhar co-founded and co-owns with his wife, Kim Jackson, has invested in a company called Stack Infrastructure, which builds data centres. That means the tech billionaire proselytising about building data centres has a financial stake in a company that builds data centres.
Sam Altman is OpenAI’s co-founder and chief executive.Credit: AP
The council’s wish list also includes “improved planning approvals to move fast both on the energy and approving of data centres”. This desire to loosen existing regulations suggests they would like something akin to the US AI Action Plan, which was unveiled by the Trump administration in July. That policy road map aims to remove roadblocks so that AI innovation in the US can sprint ahead, unfettered by pesky investigations into monopolies, third-party audits of their technology, regulations for misleading advertising or environmental protections.
The AI Action Plan paid no heed to the growing chorus asking whether an industry that has already received close to $1 trillion in investment and has a history of over-promising and under-delivering deserves to receive even more assistance and less regulation. As Matteo Wong, a tech writer for The Atlantic, wrote: “AI products remain error-prone, extremely expensive to build and (are) unproven in many business applications.”
Gary Marcus, professor emeritus of psychology and neural science at New York University, and a forceful critic of the hype surrounding AI, bluntly said:, “We have been fed a steady diet of bullshit (from AI companies) for the last several years.” He doesn’t believe that the tech lords can deliver on their AI promises and has called for regulation of AI, a move that would seem essential, given that it was less than two years ago that the board of Open AI was so concerned about Altman’s handling of AI safety issues (which address the misuse and harmful consequences of AI) that they launched a failed coup to oust him.
Whether the text and data mining exemption goes through, or Australia instead gets AI-specific legislation that provides oversight of a currently unregulated and unproven industry, will be up to our politicians. At the 2023 launch of the government’s culture policy, our music-loving prime minister gave a heartfelt speech, in which he committed to supporting the artist as worker and said: “This is about our soul … who we are and our quality of life.”
If the Albanese government ends up giving the green light to the exemption, it will allow billionaires to steal entire lifetimes of work from some of Australia’s lowest-paid workers so that tech companies, owned by some of the richest people in the world, can profit off artists’ labour in perpetuity. It’s therefore not only the words of Australian authors that are being weighed for their value in the AI debate. As we wait to see which direction the spaceship will take, we’re also about to find out if the words of our politicians have any value at all.
Melanie La’Brooy is an award-winning Australian children’s author.
The Opinion newsletter is a weekly wrap of views that will challenge, champion and inform your own. Sign up here.