The health sector has a big part to play in shaping the development of generative AI.
Spanning medical research, pharmaceutical design, and clinical trials – can Australia do more to utilise this dynamic and innovative market? Is the health sector equipped to weather these changes? How does the global backdrop of AI development encourage positive disruption?
This was the centre of Melbourne Press Club’s AI: After the Chat event held at KPMG’s Melbourne head office in April. Eleven experts from across Australia explored the impact of AI across journalism, sports, health and business development.
Cerulea Clinical Trials CEO Michelle Gallaher and CSIRO Chief Research Consultant Stefan Hajkowicz were keynote speakers who focused on the relationship and future impacts of AI in the Australian and global healthcare system.
Our own Mariam Koslay breaks down the key points from this discussion and how it can and will relate to healthcare in the future.
Generative AI Framework
Generative AI is an evolving algorithm that can learn from different AI systems to generate new content, tasks or ideas. Historically, AI was used to understand and recommend information. Now generative AI is shaping our day-to-day lives with a growing use of Chat GPT’s Open AI, Google’s Gemini and China’s Baidu Ernie.
In AI there have been traditionally two schools of contrasting approaches. Symbolic AI refers to the representation or encoding of human knowledge in the form of known facts/ rules. This is then used with accompanying data to reason, and infer other facts from data. Or Statistical AI, which uses existing data and evidence along with computational patterns to extract patterns.
CSIRO Chief Research Consultant Stephan Hajkowicz believes the ability for systems to interact with each other has contributed to the acceleration of generative AI. With close to 25 years of experience in the sector, he authored Australia’s National AI Roadmap and Ethics Framework and is currently working on the economic geography of Australia’s rapid-growth technology industry.
On January 17, 2024, the Australian Government published an interim response to safe and responsible AI consultation which included voluntary safety standards, voluntary labelling and watermarking of AI-generated materials and established an expert advisory body to support future development of guardrails.
Mr Hajkowicz shared that the adoption of generative AI in companies across Australia is motivated by a return on investment, with a focus on advanced problem-solving and developing large, multi-model and accurate data sets. While we may be getting better as a country at handling risk, we as a country are still navigating a jagged technological frontier.
Equitable AI Healthcare Systems
According to Global Market Insights, the global AI healthcare market share, which includes drug discovery, medical imaging, precision medicine and genomics coupled with personalised treatments, accounted for a market size of over $5.4B USD in 2022 and is expected to rise to $19.9b USD by 2032.
Prominent market players operating in AI Healthcare have been identified as IBM Corporation, Modernizing Medicine, NVIDIA Corporation and Sophia Genetics.
CSIRO’s Australasian e-Health Research Centre (AEHRC) report published in March of this year cited healthcare as a winner in AI, referencing technologies that reduce clinician burden, support assistive technologies in aged care, develop data harmonisation for Alzheimer’s research, and roll out therapeutic chatbot technology.
Cerulea Clinical Trials CEO Michelle Gallaher shared six ways in which AI can be applied in health care:
Specialising in gene therapy, she has witnessed the profound change of AI in producing safer design and discovery.
The unique integration of ethics and advocacy across patients, experts and technology within health the health sector will allow AI to pivot towards functional AI and equity, with the ability for the humanity of experiences to remain.
She believes healthcare AI can positively expand the tech sector, engraining a more rounded understanding of society and culture the more AI develops into the future.
If you are a company looking to adopt AI into your workplace, you can read more about Australia’s framework here.
Creating a Culture of AI
Across his work, Mr Hajkowicz encouraged Australian companies embracing generative AI to think critically about adoption, adaption and governance when developing an artificial intelligence strategy.
He cited that the greatest barrier to adopting generative AI systems successfully was the lack of attention in establishing a culture of education and understanding, to alleviate tensions between those using the AI and the AI systems itself.
In Australia, the Artificial Intelligence Ethics Principal can be listed into 8 key focus areas:
On March 13, the European Union passed the EU Artificial Intelligence Act (523-46 vote), presenting the first comprehensive regulation on AI. Establishing three risk categories: Unacceptable Risk Applications, High-Risk Applications and Low-Risk Regulations, companies in the EU can be fined up to 6 per cent of their revenue if found non-compliant.
Mr Hajkowicz cited the EU Artificial Intelligence Act as an important global benchmark to balance the growing sense of urgency and strategy.
