Urgent.News

What's breaking now, across thousands of outlets.

AI

These 3 charts show female-dominated jobs are actually the most exposed to AI

Of the top 20 occupations most at risk of being automated, 15 are dominated by women.

New analysis reveals that female-dominated jobs are among the most exposed to AI-fuelled automation, despite initial assumptions to the contrary. Women make up the majority of workers in the care sector, including child care, nursing, and aged care, leading to the belief that their jobs are largely shielded from AI impacts. However, government data from Jobs and Skills Australia shows that female-concentrated jobs are, in fact, the most at risk of being replaced by AI technologies.

This is due to the suitability of AI to alleviate human effort for repetitive routine tasks in clerical and administrative sectors, which are largely composed of women and provide nearly one in five jobs for Australian women. Fifteen of the 20 top occupations most at risk of automation are female-dominated, with five classified as highly female-dominated (at least 80% women).

These occupations include secretaries, receptionists, bookkeepers, and accounting, human resources, and payroll clerks. In contrast, only one occupation on this list is moderately male-dominated, while the remaining four are gender-balanced. The 20 occupations least exposed to automation are predominantly male-dominated, with only three female-concentrated jobs.

Job vacancy data also indicates that highly exposed jobs to automation have been experiencing a quicker decline in vacancies compared to the wider workforce. This underscores the urgency for policymakers to apply a gender lens in AI policies, ensuring that the adaptation and adoption of AI in the labor market is both adaptive and gender equitable.

Written by urgent.news from The Conversation AU's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.

Read the original at theconversation.com →

More in AI

I audited a 5.5 GB AI training dataset by downloading 0.8% of it

Two months ago a repository turned up in my corner of the internet: 3,358 stars, 744 forks, an MIT license, and a release advertised as 518,400 training samples — 5.5 GB, split across three zip parts.

  • Archive size: 5.5 GB with 518,400 training samples
  • Investigation focused on method 0.jsonl.gz shards
  • 0.81% of release obtained by requesting specific entries

More from Thursday 6 August →