- Information Media and Telecommunications most likely to be impacted.
- 14 out of the top 20 suburbs belong to Victoria.
- Policymakers and businesses need to plan for AI and automation led future.
Suburbtrends has published a new report naming the top 20 Australian suburbs most likely impacted by artificial intelligence (AI) and automation. The study aimed to gauge AI’s effect on the workforce, using the suburb as a unit of analysis.
Jobs most impacted by AI
Popular language models, Bard and ChatGPT, were used to estimate productivity gains, expressed as percentages, across 18 employment categories as defined by the Australian Bureau of Statistics (ABS). The study used the average between the two figures.
As seen below, the industries expected to be most affected by the move towards AI and automation are Information Media and Telecommunications, Professional Scientific and Technical Services, and Financial and Insurance Services.
Automation impact
Category | Productivity gain, ChatGPT | Productivity gain, Bard | Average |
---|---|---|---|
Information Media and Telecommunications | 90% | 90% | 90% |
Professional Scientific and Technical Services | 85% | 85% | 85% |
Financial and Insurance Services | 80% | 80% | 80% |
Manufacturing | 70% | 75% | 73% |
Wholesale Trade | 60% | 60% | 60% |
Health Care and Social Assistance | 50% | 65% | 58% |
Education and Training | 45% | 70% | 58% |
Retail Trade | 55% | 55% | 55% |
Administrative and Support Services | 75% | 20% | 48% |
Transport, Postal and Warehousing | 65% | 15% | 40% |
Mining | 30% | 35% | 33% |
Rental, Hiring and Real Estate Services | 35% | 25% | 30% |
Accommodation and Food Services | 15% | 45% | 30% |
Electricity, Gas, Water and Waste Services | 40% | 10% | 25% |
Agriculture, Forestry and Fishing | 20% | 30% | 25% |
Public Administration and Safety | 10% | 40% | 25% |
Construction | 25% | 15% | 20% |
Arts and Recreation Services | 5% | 15% | 10% |
Source: Suburbtrends.
After that, the estimated productivity gain refers to the total number of people employed in each job category that will be affected by AI and automation.
For instance, in the case of construction, where automation is expected to result in a 20% productivity improvement, it is assumed that 20% of people in this sector will lose their jobs — Surburbtrends emphasises that this is in the worst-case scenario.
The study then used national census data to tally the number of people employed in each of the 18 job categories in every Australian suburb. The productivity improvement percentages obtained earlier were utilised to compute the number of people impacted in each job category in each suburb.
Example suburb: Marrickville NSW
Industry | People potentially impacted |
---|---|
Professional, Scientific and Technical Services | 1,751 |
Health Care and Social Assistance | 1,006 |
Education and Training | 874 |
Financial and Insurance Services | 777 |
Information Media and Telecommunications | 642 |
Retail Trade | 586 |
Manufacturing | 431 |
Public Administration and Safety | 262 |
Accommodation and Food Services | 260 |
Transport, Postal and Warehousing | 211 |
Administrative and Support Services | 194 |
Wholesale Trade | 191 |
Construction | 151 |
Rental, Hiring and Real Estate Services | 56 |
Arts and Recreation Services | 46 |
Electricity, Gas, Water and Waste Services | 18 |
Agriculture, Forestry and Fishing | 10 |
Mining | 3 |
Source: Suburbtrends.
Finally, the Australians affected in each type in each given suburb were totalled and ranked based on the number of people involved.
Top 20 suburbs, national
Rank | State | Suburb | Total people potentially impacted |
---|---|---|---|
1 | VIC | Melbourne | 16,490 |
2 | VIC | Point Cook | 16,132 |
3 | VIC | Craigieburn | 12,667 |
4 | VIC | Berwick | 12,092 |
5 | VIC | Tarneit | 12,069 |
6 | VIC | Reservoir (Vic.) | 11,440 |
7 | VIC | Pakenham | 11,393 |
8 | NSW | Blacktown | 10,340 |
9 | VIC | Werribee | 10,313 |
10 | NSW | Castle Hill (NSW) | 10,165 |
11 | VIC | Glen Waverley | 9,942 |
12 | VIC | Richmond (Vic.) | 9,812 |
13 | NSW | Baulkham Hills | 9,403 |
14 | NSW | Dubbo | 9,209 |
15 | NSW | Parramatta | 8,998 |
16 | NSW | Port Macquarie | 8,938 |
17 | VIC | Preston (Vic.) | 8,922 |
18 | VIC | Mount Waverley | 8,813 |
19 | VIC | Rowville | 8,809 |
20 | VIC | Sunbury | 8,777 |
Source: Suburbtrends.
Victoria most likely to be affected
Victoria holds the most suburbs with the most people that may be affected by automation, with 14 out of 20 suburbs on the list belonging to the capital city. Trailing behind is New South Wales, with six suburbs on the list.
Kent Lardner, Suburbtrends founder, stresses that, “Understanding the potential ramifications of AI and automation at a granular level is crucial for planning, whether at the policy level, in business strategy, or for individual career development.”
Melbourne, Point Cook, and Craigieburn hold the top spots for most potentially impacted suburbs — Melbourne, with 16,490 individuals; Point Cook, with 16,132; and Craigieburn, with 12,667 possible job losses.
Planning for an AI-led future
“These numbers underscore the need for Australia to prepare for a future where AI and automation will undoubtedly play an increasingly pivotal role. It’s not necessarily about job losses but job transformation. Areas with higher concentrations of workers in certain industries might feel the effects more intensely, and we need to be ready for that,” Lardner adds.
According to Suburbtrends, the findings demonstrate that policymakers, businesses, and individuals must proactively plan for AI and automation’s inevitable rise.
Though the research uses the worst-case scenario to interpret productivity gains, assuming that it equates to job losses, it recognises that increased productivity does not necessarily always leads to job losses. Instead, it could lead to job transformation and redistribution, possibly producing new job types.
“We’re standing on the precipice of a significant shift in the workforce. This transformation could be as profound as the industrial revolution, but with the right preparation, Australia can navigate this change effectively,” Kent says.