Parliament hears how SA is leading the way in AI research and development
Some of the brightest minds in artificial intelligence have given South Australian politicians an expert insight into how the state is leading the way in research and development of the technology.
By Kurtis Eichler
Australian Institute for Machine Learning directors Professor Simon Lucey and Professor Anton Van den Hengel addressed the South Australian parliament’s select committee on artificial intelligence this week, the committee’s first public hearing.
They were joined by AIML’s business development advisor Dr Paul Dalby and professor of genetic epidemiology and senior researcher Professor Lyle Palmer.
The parliament formed the select committee in July to investigate how AI is being developed and applied across a range of sectors, how SA can strengthen its AI capability through its research and development sector, as well as the attraction of investment and the training and retention of AI talent.
Chaired by Member for Florey, Michael Brown, the committee comprises six members, including Labor MPs Lee Odenwalder and Erin Thompson, Liberal MPs Stephen Patterson and Josh Teague, and independent MP, Fraser Ellis.
AIML members covered a range of topics during the 90-minute public hearing, including the need to invest in AI research and development nationally and questions over government regulation of AI.
Asked by the committee about the impact of regulation on AI, Professor Palmer said there was a “very good ethical framework” around the work AIML wants to do in medicine, for example.
“Ethics is something we care very much about in medicine,” Professor Palmer said, adding that there are very good international guidelines and protocols for medical AI, and that South Australian researchers were involved in developing them.
Dalby told the committee South Australia was leading the country in funding for AI research and development, while the federal government is still behind other comparable countries.
“There wouldn’t be an Australian Institute for Machine Learning if it wasn’t for the state government,” Dalby said.
AIML was established in 2018 as a partnership between the South Australian government and the University of Adelaide; with the Department for Industry, Innovation and Science funding the institute to support local SMEs in their development of machine learning products, and to drive the adoption of machine learning technology to across local industry and the public sector.
Since then, AIML has worked with more than 35 businesses and 10 state government entities to create new AI products and use AI to automate processes to improve productivity and quality.
Professor Simon Lucey said Australia should invest in growing Australia's specialised AI strengths and capabilities.
Professor Lucey said South Australia was now a globally known brand in AI because AIML has strategically chosen to excel in specialised areas of AI—specifically computer vision—which was a strategy that should be considered nationally.
“We can’t spend the like the superpowers,” Lucey said—referring to the large investments into broadscale AI research by China and the United States—but should instead invest in growing existing specialised AI strengths and capabilities.
Brown asked the AIML members how South Australia would benefit from bolstering its hardware resources and AI capability.
In response, Professor van den Hengel said developing local and national capability in AI would be critical in controlling our own destiny.
“This is the scale of change of electricity… the only way to maintain control is to have our own capability.”
The committee has received 41 written submissions and is expected to deliver its report to parliament by the end of November.