The Automation Divide

Which Australians will AI leave behind?

Australia is in the middle of an AI revolution and millions of workers are in the crosshairs. But the fear of โ€œrobots taking jobsโ€ has always been painted in broad strokes.

The data tells a more precise story.

Each bubble is one occupation group. Further right means higher automation risk. Higher up means better pay. Hover over any bubble to see who works there and what they earn.

The first thing to notice is where the high-risk zone begins. The dashed line marks 55% automation risk. But the danger zone doesnโ€™t start at the bottom of the pay scale it cuts through the middle.

Clerical workers, sales assistants, inquiry clerks, and office administrators cluster between $700 and $1,400 a week. These are not minimum-wage jobs. They are the backbone of Australiaโ€™s white-collar working class and they are squarely in the high-risk zone.

Now look at the colour. Darker blue means more women in that workforce. The darkest bubbles sit in the high-risk zone. Clerical and administrative occupations, carers, hospitality workers overwhelmingly female, overwhelmingly exposed. The automation wave is not gender-neutral.

Look at the size of those bubbles. Each represents hundreds of thousands of Australian workers. The risk is not abstract or distant. It is already embedded in the occupational structure of the economy.

But is being โ€œat riskโ€ really that bad? Not necessarily if AI also makes your job better.

This is the distinction that most coverage misses entirely.

Being exposed to AI does not automatically mean a job is disappearing. This chart compares automation risk (horizontal axis) with augmentation potential (vertical axis).

The first thing to notice is the red quadrant. It represents the outcome many workers fear most: high automation risk with little AI assistance.

But look how empty it is. Very few occupation groups combine high replacement risk with low augmentation benefit.

That emptiness is the story. Across most occupations, automation and augmentation arrive together. AI is more likely to reshape jobs than eliminate them outright.

The real divide is elsewhere. Occupations in the upper half of the chart gain stronger augmentation benefits, while those further right face greater automation pressure.

The blue quadrant shows the most favourable position: low automation risk and high augmentation potential. Many professional occupations cluster here, suggesting they are more likely to benefit from AI than be displaced by it.

That is the structural picture. But is it showing up in the employment data yet?

Hover over any point on either line to see the exact quarterly figures.

Since 2016, employment in lower-risk occupations (green) has grown by more than 15%. Employment in higher-risk occupations (red) has grown far more slowly and the gap has been widening since 2018, well before AI became a mainstream conversation.

COVID-19 temporarily compressed the gap both groups contracted in 2020. But as the economy reopened, lower-risk occupations bounced back strongly. Higher-risk occupations recovered more slowly, never quite closing the gap that had opened before the pandemic.

Then came late 2022 when ChatGPT launched and generative AI entered everyday workplaces. Higher-risk employment flattened. The gap snapped wider. By 2025, the divergence is unambiguous: lower-risk occupations are growing, higher-risk occupations are treading water.

The divide is not a prediction. The data shows it is already happening quietly, quarter by quarter, in employment figures most Australians never see.

The AI revolution will generate enormous productivity gains. The question is not whether it will reshape the workforce it will. The question is whether Australia will design a response that shares those gains, or allow the divide to deepen along the lines it is already following: gender, occupation, and wage.

Data sources: Jobs and Skills Australia, Gen AI Capacity Study (Aug 2025); Australian Bureau of Statistics, Employee Earnings and Hours (Aug 2025); ABS Labour Force, Australia, Detailed (Feb 2026).