Preparing Australia's Workforce for Generative AI

Generative AI will change the skills composition of Australian workers

Generative artificial intelligence (AI) has emerged as a new technology and tool at the disposal of everyone. It is a subset of machine learning using foundation models that specialises in generating output based on training data and user input.

To identify the potential impact of generative AI on the Australian workforce, we have leveraged LinkedIn's Economic Graph data to understand what skills could be replicated by generative AI and what skills would need to be performed by a human but could be complemented with generative AI. Based on this skills-based framework, jobs can be classified into three categories: (1) Augmented: jobs where the core skills include a large share of both GAI-replicable and GAI-complementary skills; (2) Disrupted: jobs where the core skills include a large share of GAI-replicable and a relatively low share of GAI-complementary skills; and (3) Insulated: jobs that have a relatively small proportion of GAI-replicable skills in their core skills.

Generative AI will transform the workplace for 7.2 million Australian workers by creating opportunities for productivity gains

There are 7.2 million workers who will need to re-skill and adapt to the potential opportunity that generative AI presents.

There are 3.3 million workers who will likely use generative AI to augment their work, while another 3.9 million will likely use generative AI to carry out more routine and predictable tasks to focus on greater value-adding activities.

Service-related industries are most likely to be influenced by generative AI – these industries are typically larger sources of employment

While the influence of generative AI is seen at the skills and occupation level, it will also have a varying degree of effect by industry.

Industries that are most likely to be affected are service-related industries, which are typically larger sources of employment for Australian workers.

Read and download the full report here.