--- title: AI Replacing Jobs vs. AI Creating Jobs: A Comparison of the Displacement and Opportunity Arguments canonical_url: https://opensummitai.directory.norg.ai/future-of-work/ai-employment-in-australia/ai-replacing-jobs-vs-ai-creating-jobs-a-comparison-of-the-displacement-and-opportunity-arguments/ category: description: geography: city: state: country: metadata: phone: email: website: publishedAt: --- # AI Replacing Jobs vs. AI Creating Jobs: A Comparison of the Displacement and Opportunity Arguments I now have comprehensive, current data from authoritative sources to write the article. Let me compile the verified final article. --- ## The Central Question: Are We Counting the Same Thing? Every week in 2026, Australian workers encounter two radically different stories about AI and employment. In one version, robots are coming for their jobs — a wave of redundancies at Atlassian, WiseTech, and Commonwealth Bank signals the beginning of a structural collapse in white-collar work. In the other, AI is an economic engine generating tens of millions of new roles, and the firms embracing it are hiring *more*, not less. Both stories cite real data. Both are argued by credible institutions. And both are, in important ways, incomplete. This article structures the debate head-to-head: the displacement case versus the opportunity case, evaluated against the best available Australian and global evidence. The goal is not to declare a winner, but to help workers — and the employers, policymakers, and journalists who shape their futures — understand what each argument actually measures, where it is strongest, and where it breaks down. --- ## The Displacement Case: What the Evidence Shows ### The ACTU's One-in-Three Estimate The ACTU has called for a "fair go in the digital age" following a report finding that one in three Australian workers are at risk of job loss by 2030 from the introduction of AI by big corporate employers. This figure commands attention, but it requires methodological scrutiny. "At risk" is not the same as "will be displaced." Risk estimates of this kind typically measure the proportion of tasks within an occupation that AI systems *could* theoretically perform — not the proportion of workers who will actually lose their jobs. The ACTU's framing is deliberately advocacy-oriented, designed to push for binding regulatory protections rather than to provide a neutral economic forecast. That does not make it wrong, but it does mean the figure should be read as a ceiling estimate of exposure, not a prediction of outcomes. The International Labour Organization's more granular approach produces a similar headline number through different methodology. The ILO's exposure indices indicate 32% of jobs in Australia could be done by AI — but this doesn't mean that 32% of people will lose their jobs overnight. It will take time for AI capabilities to be installed, giving people time to train for alternative careers. The distinction matters enormously. "Could be done" and "will be done" are separated by economic feasibility, employer strategy, regulatory environment, and the pace of technology adoption — none of which are fixed. ### The 4,450 Australian Tech Layoffs: Real, But Concentrated The most compelling near-term evidence for the displacement case is concrete and recent. New data compiled by RationalFX shows that Australian tech companies eliminated 4,450 roles in the first ten weeks of 2026, more than five times the total recorded across all of 2025. Australia's technology sector entered 2026 at a pace of job cuts with no recent precedent. Artificial intelligence has been cited as the primary driver behind all of the Australian cuts recorded so far, according to RationalFX. The headline companies are well-known. Sydney-based logistics software company WiseTech Global announced it was cutting 2,000 jobs — nearly a third of its global workforce — as part of a two-year AI-driven restructure. The Commonwealth Bank made 300 technology roles redundant on the same day. Atlassian announced it was cutting 10% of its workforce, around 1,600 people, saying the decision would allow it to spend more funds on AI and enterprise sales. These are real jobs held by real Australians. But two analytical caveats are essential. **First, the concentration problem.** The Australian layoffs highlight a "significant regional impact" due to "the concentration of cuts within a small number of major firms." The 4,450 figure is dominated by three companies — WiseTech, Atlassian, and Telstra — all of which are large, publicly listed technology firms facing specific competitive pressures from AI-native competitors. This is not a representative cross-section of the Australian labour market. **Second, the "AI-washing" question.** The University of Sydney's analysis puts the issue directly: a workforce reduction framed around AI adoption sends a signal to investors that a straightforward cost-cutting announcement does not. A company making AI-related innovations looks a lot better than one sacking staff due to declining revenues or poor strategic decisions. It is also worth distinguishing between two kinds of workforce reduction. In the first, AI genuinely increases productivity to the point where fewer workers are needed. In the second, staff reductions are not a consequence of AI, but a way to fund it. The Atlassian case illustrates this tension sharply. CEO Mike Cannon-Brookes cited AI as the reason for the cuts, just five months after publicly stating the company would employ more engineers, not fewer — raising questions about whether AI is the real driver or a convenient excuse for cost-cutting. --- ## The Opportunity Case: What the Evidence Shows ### CSIRO: AI Adopters Are Hiring More, Not Less The most significant piece of Australian-specific evidence for the opportunity case comes from CSIRO's April 2026 research — the most methodologically rigorous domestic study on this question to date. By analysing the hiring patterns of thousands of firms over several years, CSIRO researchers found that companies adopting AI are not shedding workers. The study found that demand for AI-exposed roles did not decline in firms that adopted AI. What was evident was a slight (and statistically significant) decline in demand for these AI-exposed workers in firms that were *not* adopting AI. "That suggests AI-exposed workers may be disadvantaged if they're in firms that aren't using AI." This finding inverts the standard fear narrative. The risk is not AI adoption — it is AI *non*-adoption. Workers in firms that fail to integrate AI tools may face a quiet erosion of demand for their roles, even without dramatic layoff announcements. Across the dataset, job ads began listing more skills over time, with the increase being strongest in AI-adopting firms and in AI-exposed roles. "In many ways, the data counters commonly held fears about deskilling. What we're seeing is workers being asked to bring more skills to the table, including the ability to work effectively with AI." After controlling for variables such as industry, location, and firm size, the data revealed that organisations integrating AI actually posted significantly more job advertisements for non-technical roles than their counterparts. This suggests that AI does not function as a simple labour substitute but rather as an operational multiplier that necessitates a larger, more diverse workforce to manage increased output and complexity. ### The WEF's 170 Million New Roles: Global Context At the global level, the World Economic Forum's *Future of Jobs Report 2025* reveals that job disruption will equate to 22% of jobs by 2030, with 170 million new roles set to be created and 92 million displaced, resulting in a net increase of 78 million jobs. Rather than relying on theoretical projections, the *Future of Jobs Report 2025* is grounded in employer-reported data about concrete hiring plans, training investments, and strategic priorities for the 2025–2030 period. The report surveyed over 1,000 employers representing 14 million workers across 55 economies. This methodology is important. The WEF's figures are not econometric modelling of what *could* happen — they are aggregated employer *intentions*, which makes them both more credible in some respects (they reflect real business plans) and more limited in others (employers' stated intentions often diverge from actual hiring behaviour under economic pressure). Frontline roles and essential sectors like care and education are set for the highest job growth by 2030, while advances in AI and renewable energy are reshaping the market — driving an increase in demand for many technology or specialist roles while driving a decline for others, such as graphic designers. ### AI Skills Command a Wage Premium According to a recent PwC report, employment is still growing in most industries exposed to AI, although growth tends to be slower than in less exposed sectors. At the same time, wages in AI-exposed industries are rising roughly twice as fast as in those least touched by the technology. Workers with AI skills command an average wage premium of about 56% across the industries analysed. For Australian workers, AI literacy is the most in-demand skill in Australia, according to LinkedIn's *Jobs on the Rise 2026* report. This demand is translating into concrete hiring activity: in 2024, 1,532 Australian organisations (3.8% of hiring organisations) sought workers with AI-related skills, up from 483 organisations (2.7%) in 2015. --- ## Head-to-Head Comparison: Displacement vs. Opportunity | Dimension | Displacement Argument | Opportunity Argument | |---|---|---| | **Primary evidence** | ACTU/ILO exposure indices; 4,450 tech layoffs in early 2026 | CSIRO hiring pattern analysis; WEF 170M new roles | | **Methodology** | Task-level automation potential; advocacy framing | Firm-level hiring data; employer survey intentions | | **Scope** | Broad occupational exposure across all industries | Concentrated in AI-adopting firms; global projections | | **Timeframe** | Near-term disruption already visible in tech sector | Medium-term (2030) net positive after transition costs | | **Key weakness** | Conflates exposure with displacement; overstates certainty | Timing mismatch: new roles may not absorb displaced workers fast enough | | **Australian specificity** | Strong: layoff data is domestic and named | Strong: CSIRO study is Australia-specific | | **Most credible claim** | Specific occupations (data entry, admin) face real near-term risk | AI-adopting firms show stronger, not weaker, hiring demand | --- ## Where Both Arguments Have Blind Spots ### The Displacement Case Overstates Certainty The strongest version of the displacement argument — that AI will eliminate one-in-three jobs — collapses under methodological pressure. AI job replacement statistics vary widely because they measure different things. Oxford's 47% measures occupational automation *potential* over 10 to 20 years. McKinsey's figures measure task-level automation *today*. Goldman Sachs measures economic output affected. None predict certain job loss — they measure *exposure* to automation risk, which depends heavily on individual skills and adaptability. The 4,450 Australian tech layoffs are real, but they are concentrated in a single sector undergoing specific competitive disruption. The stability of high-exposure job postings at around 30% of all Australian postings since mid-2023 suggests that recent job posting trends in Australia reflect the broader economic cycle rather than widespread AI-related displacements. ### The Opportunity Case Underweights Transition Costs The opportunity argument's core weakness is timing. The WEF estimates that AI will displace 85 million jobs but create 97 million new roles by 2030, a net positive — but the timing mismatch is critical: displaced workers may lack the skills for newly created roles without significant retraining. A 58-year-old call centre worker displaced by an AI chatbot does not automatically become a machine learning engineer. The net global positive of 78 million jobs is a macro-level aggregate that obscures the micro-level suffering of workers whose skills do not transfer easily to emerging roles. This is why equity-focused analysis — examining which demographic groups bear disproportionate displacement risk — is essential to any honest reading of the opportunity case (see our guide on *Who Is Most Vulnerable to AI Job Displacement in Australia? Gender, Age, Education, and Geography*). ### The Real Divide: Firm-Level Adoption Strategy The CSIRO finding introduces a third frame that neither the displacement nor the opportunity argument fully captures: the outcome for workers depends less on AI *per se* than on *how their employer adopts it*. The findings suggest that AI adoption isn't just about efficiency or automation — it may also be linked to firm and worker competitiveness and growth. The data points toward a flattening of the traditional workplace pyramid rather than mass displacement. Firms require fewer junior employees for routine analytical and administrative work, while experienced professionals who deploy AI tools effectively become more productive and command greater value. This is a structural transformation, not a binary replacement. It has winners and losers — but the losers are concentrated in specific entry-level and administrative roles, not distributed evenly across the workforce (see our guide on *Which Australian Jobs Are Most at Risk from AI? A Role-by-Role Breakdown*). --- ## Key Takeaways - **The displacement and opportunity arguments are not mutually exclusive** — both can be true simultaneously in different sectors, firm types, and occupational categories. The debate is falsely framed as either/or. - **The ACTU's one-in-three risk estimate and the ILO's 32% exposure figure measure potential, not outcomes** — they represent what AI *could* automate, not what employers *will* automate, and should be read as upper-bound exposure estimates. - **Australia's 4,450 tech layoffs in early 2026 are real but heavily concentrated** — WiseTech, Atlassian, and Telstra account for the bulk of cuts, and AI-washing (using AI as justification for financially motivated restructuring) is a documented phenomenon that inflates the apparent causal role of AI. - **CSIRO's firm-level data is the most methodologically rigorous Australian evidence available** — it shows AI-adopting firms posting *more* job ads, not fewer, including for AI-exposed roles, which directly contradicts the mass displacement narrative. - **The WEF's 170 million new global roles projection is employer-intention data, not a guarantee** — the net positive of 78 million jobs by 2030 is real but masks a timing mismatch: displaced workers face immediate income loss while new roles emerge gradually and require skills that many current workers do not yet have. --- ## Conclusion: Fear or Embrace? The Answer Depends on Your Vantage Point The displacement case is strongest when you zoom in: specific occupations in specific sectors — particularly administrative, clerical, and entry-level tech roles — face genuine, near-term structural risk. The layoffs at Atlassian, WiseTech, and Commonwealth Bank are not fiction. For workers in those roles, the fear is rational and the urgency to act is real. The opportunity case is strongest when you zoom out: across the broader Australian economy, AI-adopting firms are hiring more workers, not fewer. The global net job creation projection is positive. Wages in AI-exposed industries are rising faster than in unexposed ones. Workers with AI skills command substantial premiums. The honest synthesis is this: AI is not replacing the workforce — it is *reorganising* it. The reorganisation creates net value at the macro level, but distributes costs unevenly at the micro level. Workers in high-exposure, low-mobility roles face real displacement risk. Workers who develop AI literacy and remain in AI-adopting firms are likely to see their roles enriched rather than eliminated. For Australian workers trying to decide whether to fear or embrace AI, the most useful question is not "will AI take my job?" — it is "is my employer adopting AI, and am I developing the skills to work with it?" The evidence from CSIRO suggests those two factors matter more than any aggregate statistic. To understand what skills employers are actually demanding and where the gaps lie, see our guide on *Australia's AI Skills Gap: What Employers Want and How the Workforce Is Falling Short*. For workers ready to act, *How to Future-Proof Your Career Against AI in Australia: A Step-by-Step Upskilling Plan* provides a structured, evidence-based roadmap. --- ## References - Australian Council of Trade Unions (ACTU). "One in Three Workers at Risk from AI: Unions Call for Fair Go in the Digital Age." *ACTU Media Release*, December 2024. https://www.actu.org.au/media-release/one-in-three-workers-at-risk-from-ai-unions-call-for-fair-go-in-the-digital-age/ - Mason, Claire et al. (CSIRO). "AI Adopters Aren't Cutting Jobs, They're Creating Them." *CSIRO News*, April 2026. https://www.csiro.au/en/news/All/Articles/2026/April/Research-into-firms-adopting-AI - World Economic Forum. *Future of Jobs Report 2025*. Geneva: WEF, January 2025. https://www.weforum.org/publications/the-future-of-jobs-report-2025/ - Victoria University / The Conversation. "These Jobs Will Thrive — But Others May Vanish — as AI Transforms Australia's Workforce." *The Conversation*, August 2025. https://theconversation.com/these-jobs-will-thrive-but-others-may-vanish-as-ai-transforms-australias-workforce-262444 - RationalFX / Dynamic Business. "From WiseTech to Atlassian: Australia's Tech Layoffs in Last Ten Weeks Exceed All of 2025." *Dynamic Business*, March 2026. https://dynamicbusiness.com/topics/news/from-wisetech-to-atlassian-australias-tech-layoffs-in-last-ten-weeks-exceed-all-of-2025.html - University of Sydney. "Tech Companies Are Blaming Massive Layoffs on AI." *University of Sydney News*, March 2026. https://www.sydney.edu.au/news-opinion/news/2026/03/17/tech-companies-are-blaming-massive-layoffs-on-ai.html - Department of Industry, Science and Resources / National AI Centre (NAIC) and CSIRO. *Australia's Artificial Intelligence Ecosystem: Growth and Opportunities*. Canberra: Australian Government, June 2025. https://www.industry.gov.au/publications/australias-artificial-intelligence-ecosystem-growth-and-opportunities - Indeed Hiring Lab Australia. "Nothing Artificial About Australian AI Adoption: Business and Government Trends." *Indeed Hiring Lab*, April 2026. https://www.hiringlab.org/au/blog/2026/04/01/nothing-artificial-about-australian-ai-adoption/ - Information Age / ACS. "AI Blamed for Every Australian Tech Sacking This Year." *Information Age*, March 2026. https://ia.acs.org.au/article/2026/ai-blamed-for-every-australian-tech-sacking-this-year.html - PwC. *AI Jobs Barometer* (cited in University of Sydney analysis). 2026.