AI in the Labor Market 2026: What the Anthropic Study Really Reveals
The Key Takeaways
- Anthropic study confirms: AI is fundamentally reshaping jobs – but not eliminating them outright
- Hardest-hit roles: Call centers (−67%), copywriting (−53%), IT consulting (−34%)
- Paradox: Highly skilled workers face the greatest exposure – but also hold the strongest advantage
- Germany: 67% of companies expect no net job losses due to AI (Bitkom 2026)
- Urgency: Professionals who delay reskilling risk falling behind by 2028
For years, executives have debated whether artificial intelligence will destroy or create jobs. The Anthropic study released in March 2026 delivers the first robust, data-driven insights – and puts speculation to rest: The labor market won’t collapse, but it will undergo radical restructuring. Anyone interpreting these findings as reassurance has missed the point entirely.
The numbers speak plainly
What Anthropic has delivered with its new labor-market analysis is no academic exercise. Between Q3 2024 and Q3 2025, job postings plummeted sharply across specific sectors: call center roles down 67%, copywriting down 53%, project management down 48%, and IT consulting down 34%. These aren’t fringe categories – they’re core knowledge-work functions.
At the same time, a February 2026 Bitkom survey found that 67% of German companies surveyed anticipate no impact on their overall headcount. Only one-fifth expect job cuts. That may sound contradictory – but it isn’t.
Siemens shows how transformation works without mass layoffs
In early 2026, Siemens announced plans to reskill approximately 6,000 employees in Germany specifically for AI-related competencies – rather than laying them off. The program covers cloud engineering, prompt design, and data-driven process optimization. The underlying logic is clear: Transitioning existing teams into new roles avoids costly recruitment cycles and retains institutional knowledge.
This example matters because it mirrors what the Anthropic study demonstrates at an industry level: Change isn’t unfolding through waves of dismissals – but via a quiet, steady shift in job profiles. The project manager of 2024 doesn’t do the same work in 2026. The copywriter of 2024 is either a prompt specialist – or unemployed. Jobs aren’t vanishing; they’re evolving.
“AI won’t create fewer jobs – it will create different ones. The question is whether we shape those transitions – or let them overwhelm us.”
Enzo Weber, Head of Forecasting and Macroeconomic Analysis, Institute for Employment Research (IAB), Nuremberg
Why highly qualified professionals are most exposed
Here lies the true surprise in the Anthropic study: It’s not factory workers or logistics staff who face the highest exposure – but highly educated knowledge workers: analysts, consultants, content creators, and project managers. The explanation is straightforward: AI models excel precisely at the tasks these groups perform daily – writing text, analyzing data, building presentations, drafting decision memos.
According to an Indeed report, demand for AI skills is growing fastest in HR, marketing, finance, and project management. In other words: The very occupational groups under the greatest pressure also hold the greatest potential – if they acquire the right capabilities. Anthropic researchers describe this as an asymmetry: high exposure coupled with high adaptability.
The counterargument: Overblown panic harms more than AI itself
Not everyone shares this sense of urgency. Enzo Weber, labor-market researcher at the Institute for Employment Research (IAB), has long warned against equating technological disruption with job destruction. His argument: Germany’s economy weathered automation in the 1990s, digitalization in the 2000s, and the platform economy in the 2010s – without sustained employment declines. Every major technology wave displaced certain tasks, yet simultaneously created entirely new occupational fields.
That’s not a flawed argument. But it underestimates velocity. Factory automation unfolded over decades. The AI transformation is happening in quarters. Just under three years separate the GPT-4 launch in March 2023 from the sharp drop in copywriting job postings. This pace overwhelms upskilling systems designed around annual cycles.
What decision-makers must do – right now
The Anthropic study offers no dystopian forecast – but it does deliver a clear mandate for action. Five concrete measures emerge directly from the data:
● Conduct a skills audit: Which roles within your organization can be augmented – or automated – by AI tools? The answer is almost never “all” and almost never “none.”
● Prioritize upskilling over hiring: Reskilling current employees costs less than recruiting scarce AI specialists from a near-empty talent market. Siemens leads the way – but mid-sized firms can begin with smaller-scale programs.
● Run AI pilots in live operations – not innovation labs: Embed an AI assistant in sales, automate reporting in controlling, deploy prompt-based draft generation in communications.
● Define new roles: Prompt engineer, AI trainer, data curator – these titles barely existed two years ago. When writing job descriptions today, don’t just reflect the present state – anticipate the target state 18 months out.
● Embrace speed: Transformation won’t wait for your next strategic planning cycle. Think quarterly initiatives – not three-year roadmaps.
Conclusion: The labor market won’t shrink – it will change
At its core, the Anthropic study reveals this: The 2026 labor market has far less to do with job cuts than with job reconfiguration. Companies that grasp this early – and develop their teams accordingly – will dominate the next decade. Those who wait until their own sector suffers the same steep declines seen in copywriting or IT consulting have already missed the window.
According to recent surveys, 59% of German workers fear AI will cost them their jobs. That anxiety is understandable – but it should spur action, not paralysis. Because the data also show this: Building AI competencies doesn’t just offer protection – it makes professionals more sought-after than ever before.
Frequently Asked Questions
Will AI eliminate large numbers of jobs in Germany?
The Anthropic study shows AI isn’t wiping out jobs overnight – but it is transforming them fundamentally. Sixty-seven percent of German companies expect no net job losses. Yet task profiles are shifting dramatically – especially in knowledge-intensive fields like consulting, copywriting, and project management.
Which professions are most affected?
Call centers (−67% in job postings), copywriting (−53%), project management (−48%), IT consulting (−34%). Common denominator: High volumes of structured information processing.
Are highly qualified professionals shielded from AI?
Quite the opposite – they’re most exposed. But they also possess the strongest foundation to harness AI as a tool rather than be replaced by it. The decisive factor is willingness to reskill.
What should companies do now?
Three concrete steps: First, conduct a skills audit (which tasks are automatable?). Second, launch mandatory upskilling programs. Third, define new roles requiring AI proficiency.
Is the Anthropic study reliable?
The study draws on job-platform data and corporate surveys. It’s the first large-scale investigation offering credible, quantified insights into the AI-driven labor-market shift. Its methodology is transparently documented – and its findings align with parallel studies from Bitkom and McKinsey.
Further Reading in the MBF Media Network
- Agentic AI in Mid-Sized Companies – How Autonomous AI Agents Are Transforming Business Processes (MyBusinessFuture)
- Workforce Analytics – Data-Driven People Decisions, Not Gut Feelings (MyBusinessFuture)
- AI Cloud Costs Spiraling Out of Control – Why GPU Workloads Are Blowing Up IT Budgets (cloudmagazin)
- AI Governance at the Board Level – Why CEOs Cannot Delegate AI Responsibility (Digital Chiefs)
Header Image Source: Matheus Bertelli / Pexels
