The Australian AI Strategy
Artificial Intelligence Ethics, Strategies and Roadmap
Australia has had some dire issues lately with forest fires. That remains to be seen, in Australia artificial intelligence is both engaged in activity devastating to the environment and possibly to assist in future challenges the country is facing due to the changing environment. In this article I will touch on AI ethics, strategies and the new AI roadmap released in 2019.
AI Ethics in Australia
Will the AI system you are developing or implementing be used to make decisions or in other ways have a significant impact (positive or negative) on people (including marginalised groups), the environment or society?
- Achieve better outcomes
- Reduce the risk of negative impact
- Practice the highest standards of ethical business and good governance
- Human, social and environmental wellbeing: Throughout their lifecycle, AI systems should benefit individuals, society and the environment.
- Human-centred values: Throughout their lifecycle, AI systems should respect human rights, diversity, and the autonomy of individuals.
- Fairness: Throughout their lifecycle, AI systems should be inclusive and accessible, and should not involve or result in unfair discrimination against individuals, communities or groups.
- Privacy protection and security: Throughout their lifecycle, AI systems should respect and uphold privacy rights and data protection, and ensure the security of data.
- Reliability and safety: Throughout their lifecycle, AI systems should reliably operate in accordance with their intended purpose.
- Transparency and explainability: There should be transparency and responsible disclosure to ensure people know when they are being significantly impacted by an AI system, and can find out when an AI system is engaging with them.
- Contestability: When an AI system significantly impacts a person, community, group or environment, there should be a timely process to allow people to challenge the use or output of the AI system.
- Accountability: Those responsible for the different phases of the AI system lifecycle should be identifiable and accountable for the outcomes of the AI systems, and human oversight of AI systems should be enabled.
The principles are voluntary. They are aspirational and intended to complement–not substitute–existing AI related regulations.
AI Strategy in Australia
There has been several different initiatives relating to artificial intelligence.
- AI and automation are already considered under the national Innovation Strategy.
- The Victorian All-Party Parliamentary Group on Artificial Intelligence (VAPPGAI) was launched in Australia at the Parliament House on March 7, 2018. The launch was led by the Minister for Innovation Philip Dalidakis and Shadow Minister David Southwick. The AI parliamentary group will discuss the critical and transformative nature of AI from all sides of the political spectrum.
- The Australian Government unveiled a new Digital Economy Strategy on September 19, 2017, with the goal of launching the strategy in 2018. The strategy lays out a roadmap for the government, the private sector and the community to work together to discover new sources of growth, develop world-leading digital business, support lifelong learning, and address the ‘digital divide’ in skills and confidence. Artificial intelligence is one of the emerging technologies mentioned.
The Australian Roadmap for AI
The Artificial Intelligence Roadmap was published by the Australian Government in November 2019, and codeveloped by CSIRO’s Data61 and the Department of Industry, Innovation and Science.
This study was funded by the Australian Government. The roadmap identifies three high potential areas of AI specialisation for Australia.
It lists three facts in the beginning:
- Digital technologies, including AI, are potentially worth AU$315 billion to the Australian economy by 2028 and AI could be worth AU$22.17 trillion to the global economy by 2030.
- Australian industry needs up to 161,000 new specialist AI workers by 2030 in machine learning, computer vision, natural language processing and other AI technologies.
- The Australian information, communications and technology (ICT) sector employs 663,100 workers in fields related to AI. This will grow to 758,700 workers by 2023 at a rate of 20,000 additional workers per year. Today 66,000 ICT workers live in remote and regional areas.
- An AI system for detecting skin cancer (which could become a smartphone app) performed “on par” with 21 certified dermatologists. In Australia 13,280 new cases of melanoma were diagnosed in 2016 and 1,770 people died from the disease. Earlier diagnosis enabled by AI will save lives.
- An on-farm agricultural robot Agbot II developed by the Queensland University of Technology could save Australia’s farm sector AU$1.3 billion per year by automating weed removal and improving agricultural productivity.
- AI could help reduce the 1,137 deaths/year and 57,000 injuries/year on Australian roads. Autonomous emergency braking combined with forward warning collision systems decreased front-to- rear injury accidents on US highways by 56%. When combined with lane departure correction these semi-automated technologies could prevent or reduce the severity of 1.3 million car crashes on US roads each year.
- Mining operations in the Pilbara region of Western Australia are amongst the world’s most automated. Fortescue operates 112 driverless trucks with a 30% productivity gain. BHP has 50 autonomous trucks at its Jimblebar site and 20 autonomous drills state — wide. Launched in 2008 Rio Tinto’s mine — of — the-future has 140 automated trucks, over 11 automated drills, and 60% of train kilometres are in autonomous mode.
- In 2018 there were 1,465 start-ups in Australia, 21% of which consider AI relevant to their products or services. This increased 6% from 2017 . Through this study we surveyed 72 AI start-up companies of which 84% report to be adapting or developing novel algorithms rather than simply using AI toolboxes.
- AI is being used to help maintain Sydney Harbour Bridge. Data61 placed 2400 sensors on the bridge which monitor 800 steel and concrete supports under the roadway. Data generated by these sensors is analysed using machine-learning and predictive analytics to identify priority locations for proactive maintenance. This keeps costs down and minimises traffic disruptions.
- Drones are already being used to perform rescues and spot sharks in the Australian surf. On 18 January 2018 two teenage boys in the surf 700m from land (in a 3m swell) near Lennox Head, NSW were rescued by an airborne drone which dropped an inflatable floatation device to them. Rescue drones can be equipped with AI systems which can automatically detect and help people in need. Machine learning and computer vision systems are being developed that will automatically identify rip currents so beachgoers can be warned.
This is how the report presents AI domains:
Still the funding in Australia has varied.
So it remains to be seen whether there will be a more consistent funding for AI.
They conclude by saying the following:
“Humanity is in the process of building machines which have the ability to learn themselves without explicit human guidance. AI is already coding its own AI. We are witnessing the convergence of powerful technologies with unknown future possibilities. That’s why the world is waking up to AI and that’s why this document has been written. The ability of the machine to learn represents a new way to solve our most challenging problems. This can boost the productivity of Australian industry and improve the quality of life enjoyed by current and future generations of Australians.”