Neural Notes: CSIRO’s AI blueprint, Leigh’s big data vision and Adobe’s AI advancements

Welcome back to Neural Notes, a weekly column where I explore some of the most interesting — and sometimes weird — news happening in the Australian ecosystem.

This week: Andrew Leigh speaks about how AI and big data are affecting government operations, David Pocock proposes the establishment of the Select Committee on Adopting Artificial Intelligence and the CSIRO takes a look at foundational models in AI. Meanwhile, in Las Vegas, Adobe unveils more AI products.

CSIRO report on Australia’s AI capabilities

The CSIRO, Australia’s national science agency, has released a report detailing the country’s potential to develop sovereign capabilities in foundational AI technologies. These include high-profile AI products like OpenAI’s ChatGPT, Microsoft’s Copilot, and Google’s Gemini — which are built on foundation models that can process vast datasets and carry out a wide array of complex tasks.

According to the report, over 125 foundation models have been developed globally, with a significant majority originating from the United States, China, and Europe.

The CSIRO identifies an opportunity for Australia to develop its own models, tailored to the unique needs of its industries and services.

The report’s lead author, Dr Stefan Hajkowicz, alongside professor Elanor Huntington, CSIRO’s Digital, National Facilities & Collections executive director, highlighted the importance of Australia taking a unified approach to its AI development.

The report suggests that homegrown models could have wide-ranging applications, including in healthcare, where they could analyse complex data in patient records to potentially reduce medical misdiagnoses.

“We’ve all been impressed by the way these models can write a wedding speech or a poem. But the speed, power, and colossal scale of the data analysis they can achieve has the potential to help us solve our greatest challenges, boost productivity and save lives,” Dr Hajkowicz said.

“A foundation model for healthcare for example could help us untangle complex, hidden relationships in patients’ health records, helping us reduce the 140,000+ medical misdiagnoses in Australia each year due to human error.”

The report notes the benefits of leveraging existing models but also caution against the reliance on foreign technology due to potential security risks and cultural misalignments.

“While there are significant benefits to fine-tuning existing models in terms of cost and the speed of innovation, using foreign models poses security and reliability risks,” professor Huntington said.

“It may also result in tools that aren’t culturally appropriate in an Australian context, or that don’t realise the benefits for our workers that we want to see.”

The document calls for action in several areas to advance Australia’s position in AI, including the development of AI models for public sector use, enhancing computing capabilities, and promoting international cooperation in AI research and development.

While recognising the opportunities presented by foundation models, the report also acknowledges the challenges and risks, advocating for a careful and considered approach to AI adoption in Australia.

This wasn’t the only AI-focused report we saw from the CSIRO this week. It also released a 48-page report on AI Trends For Healthcare.

Andrew Leigh’s vision for Big Data

Source: Andrew Leigh

At the 10th Annual Australian Government Data Summit in Canberra, Assistant Minister for Competition, Charities, Treasury and Employment Andrew Leigh delivered a vision of how AI and big data are reshaping government operations and policy-making. His address painted a picture of a future where data and AI drive innovation, efficiency, and ethical governance.

Leigh utilised case study examples of AI’s impact, from updating the Australian and New Zealand Standard Classification of Occupations (ANZSCO) with generative AI to leveraging big data for economic insights, highlighting the Australian Bureau of Statistics’ (ABS) role as a pioneer in the methodological evolution of national statistics.

“Throughout, ABS data scientists were clear about the purpose of the exercise – to use generative AI to support – not replace – human analysts. All outputs were evaluated against four criteria: quality, ethics, legality, and security,” Leigh said.

“This meant that AI outputs were correct, maintained privacy, did not break intellectual property laws and vitally, maintained the place of humans at the centre of decision-making.

“This delivers on our Government’s commitment to fostering an innovative culture in the Australian Public Service, while managing the risks of emerging technologies.”

Leigh showcased the ABS’s utilisation of supermarket scanner data to monitor household inflation, demonstrating the capacity of big data to inform economic policy. He advocated for a balanced view on AI and big data, emphasising their utility in improving administrative practices without compromising ethical standards or human oversight.

“The current revolutions in AI and big data in national statistics are not simply good just because they use new, cutting-edge technologies. They matter because they offer the Australian Government a way of improving our administrative practices, and therefore the way we deliver for all citizens,” Leigh said.

“Big data and generative AI offer opportunities across Government to deliver superior outcomes for Australians. Our government remains committed to exploring these technologies within a clear ethical framework. As The Knibb knew a century ago: better data allows us to better serve the nation.”

Senate motion establishes AI Committee

A recent Senate motion has led to the creation of the Select Committee on Adopting Artificial Intelligence, spearheaded by Senators Ayers, Shoebridge, and David Pocock. This committee is charged with exploring the implications of AI technology adoption in Australia, scrutinising both the potential benefits and risks associated with AI developments.

The motion outlines a comprehensive agenda, including examining trends in AI, assessing risks such as bias and discrimination, and identifying avenues to leverage AI for societal benefits like health and environmental management.

The committee aims to present its findings by September 19, contributing to a nuanced understanding of AI’s role in Australia’s future.

Adobe enhances AI capabilities at yearly summit

Image: Adobe

Adobe has unveiled new generative AI features at its annual summit in Las Vegas. The company says that the new offerings will help bolster its Experience Cloud and Creative Cloud solutions, aiming to streamline content creation and personalisation for brands.

Anil Chakravarthy, Adobe’s Digital Experience Business president, highlighted the role of AI in reshaping customer interactions:

“Generative AI enables a fundamental shift in the relationship between brands and their customers… Adobe is leading the way in making this a reality with powerful generative AI integrated into existing workflows,” Chakravarthy said.

Key developments include Adobe GenStudio for campaign and content management, alongside an AI assistant designed to automate tasks and support customer journey mapping. Adobe also introduced AI enhancements in applications such as Adobe Experience Manager and Adobe Content Analytics, aimed at personalising marketing assets and aligning content with performance goals.

The Adobe Experience Cloud has also been updated to improve customer experience delivery, featuring new tools for unified experimentation and real-time adaptation to user behavior changes. Adobe Journey Optimizer and its B2B edition offer targeted strategies for sales and marketing teams, leveraging Adobe’s data and AI to personalise interactions.

Other AI news this week

• Australian startup Leonardo.ai — which raised $47 million back in December — has been used to create nonconsensual sexual images of celebrities
• Amazon has concluded a US$4 billion investment into Anthropic
• The UN General Assembly adopts landmark resolution on AI that aims to promote safe and ethical development across the globe
• Databricks releases the most powerful open-source AI model to date.