Generative AI is reshaping work faster than most people realise - but not evenly. Nearly one in five Australian jobs faces high AI exposure, and whether that means opportunity or risk depends largely on what you do and how much you earn. These five charts explore that divide.
Around 36.9% of Australian working-age adults use generative AI tools - ahead of the US (28.3%) but well behind Norway (46.4%), Ireland (44.6%), and France (44%). Singapore and the UAE lead globally at over 60%.
Australia is highlighted in orange.
Source: Ritchie, H., Mathieu, E., & Roser, M. (2025). Estimated share of working-age adults who use generative AI. Our World in Data. Based on Stanford AI Index 2025. Australia highlighted in orange.
Not all AI exposure is the same. Augmentation means AI helps you do your job better; automation means AI could do the job instead of you. The chart below shows how that split plays out across Australia’s major industries.
Toggle the legend to compare categories. Finance leads on augmentation; transport and retail carry the most automation risk.
Source: Jobs and Skills Australia. (2025). Our Gen AI Transition: Industry Exposures & Adoption. Australian Government. https://www.jobsandskills.gov.au/studies/generative-artificial-intelligence-capacity-study. Figures represent estimated share of workers in each industry by primary AI exposure type, based on task-level analysis using the ILO/Felten framework adapted to ANZSCO v1.3.
The industries benefiting most from AI augmentation - finance, professional services, information - also tend to be the higher-paying ones. That’s not a coincidence.
Here’s what makes AI different from previous waves of automation: it’s the higher-paid occupations that face the most exposure - but mostly in a good way. Professionals and managers sit top-left (high augmentation, low automation risk). Labourers and machinery operators sit bottom-right. Clerical workers are the uncomfortable middle ground - high automation risk despite being white-collar.
Bubble size = number of workers. Colour = pay level. Hover for details.
Source: Jobs and Skills Australia. (2025). Our Gen AI Transition: Exposures, Adaptation, Dynamism. Scores based on ILO/Felten (2021) framework applied to ANZSCO v1.3 task-level data. Employment: ABS Labour Force Survey (2024). Earnings: ABS Employee Earnings and Hours (2024).
Education level tells a similar story to pay. Higher qualifications correlate with more augmentation benefit and less automation risk. But workers with Certificate III/II - around 1.85 million Australians, including many tradespeople and admin workers - sit in a squeeze zone: moderate augmentation potential but automation risk nearly as high as those with no post-school qualification.
Orange = augmentation potential, dark blue = automation risk. Grey bars show number of workers at each level.
Source: Jobs and Skills Australia. (2025). Our Gen AI Transition. Mean augmentability and automatability scores aggregated by qualification level. Worker counts: ABS Education and Work Survey (2024).
The previous charts show potential exposure. This final chart shows what’s actually happening among workers who are already deep into using AI - software developers - and it reveals a concerning pattern.
Since 2022, AI tool usage among developers has nearly doubled. Daily use has jumped from 12% to 51%. But over the same period, trust in AI-generated output has dropped from 62% to just 38% - an all-time low according to the 2025 Stack Overflow survey. People are using it more and trusting it less.
Hover over each point for exact figures. Click the legend to show/hide lines.
Source: Stack Overflow. (2024). 2024 Developer Survey. https://survey.stackoverflow.co/2024/; Stack Overflow. (2025). 2025 Developer Survey. https://survey.stackoverflow.co/2025/. “Trust in AI code accuracy” derived from reported confidence metrics across survey years.
If this trust gap widens as AI moves into healthcare, law, and finance - where errors cost more than broken code - it becomes a serious concern for Australia’s workforce and policymakers alike.
“While AI job displacement dominates headlines, Australian data reveals a quieter, more consequential story: a growing divide between workers being augmented and workers being automated.”
Australian finance and professional services workers are gaining AI as a superpower. Meanwhile, machinery operators, retail workers, and labourers face increasing automation of the very tasks that define their roles. The divide maps almost perfectly onto income and education - reinforcing existing inequality rather than disrupting it.
What makes this story timely is Chart 5: the professionals adopting AI fastest are also trusting it least. A workforce embracing tools it doesn’t fully trust, across an economy with an uneven safety net, is a recipe for turbulence. This pitch proposes a data-driven visual story for readers who want to understand - not fear - what AI means for Australia’s working future.
Australian Bureau of Statistics. (2024). Labour force, Australia (Cat. No. 6202.0). ABS. https://www.abs.gov.au
Australian Bureau of Statistics. (2024). Employee earnings and hours, Australia (Cat. No. 6306.0). ABS. https://www.abs.gov.au
Australian Bureau of Statistics. (2024). Education and work, Australia (Cat. No. 6227.0). ABS. https://www.abs.gov.au
Felten, E., Raj, M., & Seamans, R. (2021). Occupational, industry, and geographic exposure to artificial intelligence: A novel dataset and its potential uses. Strategic Management Journal, 42(12), 2195–2217. https://doi.org/10.1002/smj.3286
Gmyrek, P., Berg, J., & Bescond, D. (2023). Generative AI and jobs: A global analysis of potential effects on job quantity and quality. International Labour Organization. https://doi.org/10.54394/FHEM8239
Jobs and Skills Australia. (2025). Our Gen AI Transition: Exposures, adaptation, dynamism. Australian Government. https://www.jobsandskills.gov.au/studies/generative-artificial-intelligence-capacity-study
Jobs and Skills Australia. (2025). Our Gen AI Transition: Industry exposures & adoption. Australian Government. https://www.jobsandskills.gov.au/studies/generative-artificial-intelligence-capacity-study/industry-data-on-ai-exposure
Ritchie, H., Mathieu, E., & Roser, M. (2025). Estimated share of working-age adults who use generative AI. Our World in Data. https://ourworldindata.org/grapher/estimated-share-people-generative-ai
Stack Overflow. (2024). 2024 Developer Survey. https://survey.stackoverflow.co/2024/
Stack Overflow. (2025). 2025 Developer Survey. https://survey.stackoverflow.co/2025/
I used ChatGPT occasionally during this project to help troubleshoot R errors and get suggestions on chart layout. All data collection, analysis, visualisation decisions, and written content are my own work.
# All code is embedded in the chunks above.
# To view with echo = TRUE, change the setup chunk option to echo = TRUE and re-knit.