Standards Australia has commenced a consultation to carry forward standards-building efforts informing the development of artificial intelligence (AI) solutions in Australia.

Key takeouts

Governments around the world are considering how to best regulate and capitalise on AI technology while limiting the risk of poor outcomes. Many have commenced efforts to develop standards.

Standards Australia, the peak standards-setting organisation for Australia, has released its first discussion paper on how it should approach standards building for AI.

Standards Australia is accepting public submissions until 31 July on the role it should play in shaping standards for AI both nationally and globally.

Standards Australia, the country's peak standards development body, has recently commenced consultation efforts surrounding the development of standards for artificial intelligence (AI). To frame discussions, Standards Australia released a discussion paper canvassing the importance of AI and summarising standards-building efforts both in other jurisdictions and at the global level. The body is now requesting submissions on how it should approach the development of standards in this space.

The discussion paper notes that cohesive international standards are particularly valuable in the ICT arena. Particular areas of interest for Standards Australia in the AI standards-setting realm include big data, machine learning, risk management, avoiding bias, ethical concerns, and governance within organisations. Given the efforts underway within the International Standards Organisation (ISO) and International Electrotechnical Commission (IEC) to develop international standards, Standards Australia is eager for Australian stakeholders to help shape Australian and, in turn, global standards.

As a first step in this process, Standards Australia is seeking submissions from the public on how Australians see the role of AI, what particular strengths Australia might have in the AI space, the role that Australian stakeholders see standards playing with respect to AI, and whether Australian standards should be designed with a national or international focus. While standards are not binding per se, it is important to remember they are commonly used as benchmarks in legal frameworks and contracts.

The Australian Government is taking a particular interest in the regulatory framework surrounding AI. In the May 2018 Federal Budget, the Government provided for the delivery of the Standards Australia report, alongside an AI Ethics Framework and an AI Technology Roadmap - both delivered by Data61 within the CSIRO. Australia's ethics framework is discussed further in Beyond Asimov's Three Laws: A new ethical framework for AI developers. The Australian Human Rights Commission is also conducting a public consultation on AI governance.

Standards Australia is holding consultation forums throughout July with final submissions due by the end of the month, and an AI Standards Roadmap is scheduled to be delivered to the Commonwealth Government in September 2019. With the economic and social impact of AI only just beginning, and nascent regulatory efforts underway around the world, the delicate balancing act of innovation and regulation will no doubt be underway for some time. Almost certainly, it will have far-reaching effects on the ultimate impacts of AI both domestically and globally.