A national-first royal commission will investigate the dangers and benefits of artificial intelligence.
South Australian Premier Peter Malinauskas announced the inquiry after returning from the United States, where he held meetings with AI companies including OpenAI, Anthropic, and Apple.
The terms of reference will be unveiled in four to six weeks, with an international search for three royal commissioners who will start work in October and report to the government by July 1.
“We want people with public policy experience thinking through issues of work and labour, but also people who obviously have a familiarity with technology and how it will inform the way people work into the future,” Malinauskas told reporters on Aug. 10.
“Equally as important as the people who lead the royal commission will be the terms of reference.”
The premier had not yet spoken to Prime Minister Anthony Albanese about the inquiry.
The cost would be about $3 million, but it would be “chicken feed” compared to the cost to society of the complex questions that AI raised, Malinauskas said.
“It is absolutely clear that the use of artificial intelligence is an exciting opportunity to advance the cause of humanity,” he said.
“But it is also true that, unchecked, artificial intelligence does represent a material risk to the way our society operates, and a risk to the future of work if not thought through carefully.”
Seventy percent of the meetings Malinauskas had in the United States involved AI and its infrastructure.
“But data centres are simply an infrastructure play,” he said.
“The main game regarding artificial intelligence is … how the technology is deployed in our society.”
Malinauskas said only governments had the tools and resources to ensure human wellbeing was at the centre of AI-related technologies and industries.
“We can choose to set the guardrails and the policy frameworks to ensure that we use AI in a way that puts people first, rather than people finding their world changing around them without their interests being heard,” he said.
Malinauskas said he had “started fiddling around” with Claude and ChatGPT.
“It’s impossible to not be utterly impressed by its capability,” he said.
Malinauskas visited OpenAI’s headquarters in San Francisco where he signed a memorandum of understanding (MoU) with its co-founder and President Greg Brockman.
OpenAI is the developer of ChatGPT, the world’s most popular generative AI chatbot.
He also met with Anthropic co-founder Tom Brown.
The federal government in April announced an MoU with Anthropic, creator of the Claude AI assistant.
Prime Minister Albanese has flagged plans to introduce legal requirements governing AI use and data centres.
He will convene a meeting of the nation’s leaders later in August to discuss data centres and reach agreement on uniform standards.
Malinauskas said he hated the idea of a world where artificial intelligence usurped the creative industries and the beauty and power of art.
“I hate the idea of kids graduating from high school and they’ve learned how to use ChatGPT, but not how to read, write, and do arithmetic … or how to communicate with one another,” he said.
“We’ve got to get the balance right here.”
The premier, whose government pioneered a social media ban for under-16s, said the issue of AI was more consequential.
“Frankly, I think AI gone bad will make social media look like a walk in the park,” he said.
