%PDF-1.4 %âãÏÓ 1 0 obj << /Type /Catalog /Pages 2 0 R >> endobj 2 0 obj << /Type /Pages /Count 6 /Kids [5 0 R 7 0 R 9 0 R 11 0 R 13 0 R 15 0 R] >> endobj 3 0 obj << /Type /Font /Subtype /Type1 /BaseFont /Helvetica >> endobj 4 0 obj << /Type /Font /Subtype /Type1 /BaseFont /Helvetica-Bold >> endobj 5 0 obj << /Type /Page /Parent 2 0 R /MediaBox [0 0 595.28 841.89] /Resources << /Font << /F1 3 0 R /F2 4 0 R >> >> /Contents 6 0 R >> endobj 6 0 obj << /Length 5286 >> stream BT /F2 22 Tf 0.06 0.08 0.12 rg 1 0 0 1 46 789.89 Tm (AI Is Hitting White Collar Jobs First as New) Tj ET BT /F2 22 Tf 0.06 0.08 0.12 rg 1 0 0 1 46 762.89 Tm (Research Reveals the Real Automation Frontline) Tj ET BT /F2 11 Tf 0.72 0.14 0.18 rg 1 0 0 1 46 725.89 Tm (TechRounder PDF Edition) Tj ET BT /F1 9.5 Tf 0.36 0.39 0.46 rg 1 0 0 1 46 709.89 Tm (Live article:) Tj ET BT /F1 9.5 Tf 0.36 0.39 0.46 rg 1 0 0 1 46 697.39 Tm (https://www.techrounder.com/ai/ai-is-hitting-white-collar-jobs-first-as-new-research-reveals-the-real-automation-f) Tj ET BT /F1 9.5 Tf 0.36 0.39 0.46 rg 1 0 0 1 46 684.89 Tm (rontline/) Tj ET q 0.82 0.85 0.9 RG 1 w 46 666.39 m 549.28 666.39 l S Q BT /F1 10 Tf 0.24 0.27 0.32 rg 1 0 0 1 46 654.39 Tm (By Vipin PG | Published March 9, 2026 | Updated March 9, 2026 | Format: Analysis | 7 min read) Tj ET BT /F2 13 Tf 0.72 0.14 0.18 rg 1 0 0 1 46 631.39 Tm (In brief) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 611.39 Tm (When people imagine automation taking jobs, they usually picture factory robots replacing assembly) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 596.39 Tm (line workers. That image has shaped the debate for decades.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 571.39 Tm (When people imagine automation taking jobs, they usually picture factory robots replacing assembly) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 556.39 Tm (line workers. That image has shaped the debate for decades. But the newest evidence suggests the) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 541.39 Tm (first serious pressure from generative AI is showing up somewhere very different: inside offices.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 519.39 Tm (A recent study released by Anthropic on March 5, 2026 offers one of the clearest real-world) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 504.39 Tm (snapshots so far of where AI is already doing meaningful work. The surprising part is not just which) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 489.39 Tm (tasks AI can perform. It's the type of workers those tasks belong to.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 467.39 Tm (Many of the most exposed roles are not manual jobs at all. They are white-collar positions built around) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 452.39 Tm (writing, documentation, digital coordination, analysis, and software development. In other words, the) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 437.39 Tm (kind of work that lives inside documents, spreadsheets, help desks, dashboards, and code editors.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 415.39 Tm (That shift challenges a long-standing assumption about automation: that it would hit physical labor first.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 400.39 Tm (Instead, the earliest measurable AI pressure appears to be forming in knowledge work.) Tj ET BT /F2 15 Tf 0.08 0.1 0.14 rg 1 0 0 1 46 372.39 Tm (Measuring Where AI Is Already Working) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 348.39 Tm (What makes this study stand out is the way Anthropic approached the problem.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 326.39 Tm (For years, most research focused on potential exposure - what AI might be capable of if models keep) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 311.39 Tm (improving. That type of analysis tends to be speculative. It assumes a future where AI systems are) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 296.39 Tm (widely deployed across industries.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 274.39 Tm (Anthropic took a different route.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 252.39 Tm (The researchers introduced a metric they call "observed exposure." Instead of asking what AI could) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 237.39 Tm (theoretically do, they examined millions of real interactions with the Claude AI system to see what tasks) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 222.39 Tm (people are already using AI for in professional settings.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 200.39 Tm (The metric blends two things:) Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 46 178.39 Tm (- Task-level estimates of what language models are capable of performing) Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 46 161.59 Tm (- Actual usage patterns showing how AI is being applied in real workflows) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 144.79 Tm (Importantly, the analysis focuses on work-related automation, not casual use. Writing an email draft or) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 129.79 Tm (summarizing a report for work counts. Asking AI to plan a vacation does not.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 107.79 Tm (The goal is simple: measure present-day task penetration, not hypothetical future disruption.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 85.79 Tm (And the results paint an interesting picture.) Tj ET q 0.86 0.88 0.92 RG 1 w 46 42 m 549.28 42 l S Q BT /F1 8.4 Tf 0.42 0.45 0.5 rg 1 0 0 1 46 30 Tm (TechRounder | Page 1 of 6) Tj ET BT /F1 7.2 Tf 0.42 0.45 0.5 rg 1 0 0 1 46 19 Tm (https://www.techrounder.com/pdf/blog/ai-is-hitting-white-collar-jobs-first-as-new-research-reveals-the-real-automation-frontline.pdf) Tj ET endstream endobj 7 0 obj << /Type /Page /Parent 2 0 R /MediaBox [0 0 595.28 841.89] /Resources << /Font << /F1 3 0 R /F2 4 0 R >> >> /Contents 8 0 R >> endobj 8 0 obj << /Length 4658 >> stream BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 789.89 Tm (Even in occupations where AI could theoretically assist with most tasks, real adoption remains far) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 774.89 Tm (lower. For example, in computer and mathematics roles, models may be capable of handling about 94) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 759.89 Tm (percent of tasks in theory, yet Claude's current measured coverage sits around 33 percent.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 737.89 Tm (So AI isn't doing everything yet. But it is clearly doing something.) Tj ET BT /F2 15 Tf 0.08 0.1 0.14 rg 1 0 0 1 46 709.89 Tm (The Jobs With the Highest AI Exposure) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 685.89 Tm (The list of occupations with the highest observed exposure reveals a clear pattern.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 663.89 Tm (These roles share three characteristics: they are digital, language-heavy, and structured.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 641.89 Tm (Anthropic's top ten most exposed jobs include:) Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 46 619.89 Tm (- Computer programmers - 74.5% observed exposure) Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 46 603.09 Tm (- Customer service representatives - 70.1%) Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 46 586.29 Tm (- Data entry keyers - 67.1%) Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 46 569.49 Tm (- Medical record specialists - 66.7%) Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 46 552.69 Tm (- Market research analysts and marketing specialists - 64.8%) Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 46 535.89 Tm (- Wholesale and manufacturing sales representatives) Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 46 519.09 Tm (- Financial and investment analysts) Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 46 502.29 Tm (- Software quality assurance analysts and testers) Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 46 485.49 Tm (- Information security analysts) Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 46 468.69 Tm (- Computer user support specialists) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 451.89 Tm (None of these jobs involve repetitive physical motion. They revolve around text, documentation,) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 436.89 Tm (analysis, and digital systems.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 414.89 Tm (That makes them a natural fit for large language models.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 392.89 Tm (Tasks such as summarizing documents, drafting reports, translating technical findings into plain) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 377.89 Tm (language, classifying information, and generating structured content are exactly what these systems) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 362.89 Tm (are designed to handle.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 340.89 Tm (And that alignment shows up clearly in how companies are already using them.) Tj ET BT /F2 15 Tf 0.08 0.1 0.14 rg 1 0 0 1 46 312.89 Tm (What AI Is Actually Automating Inside These Jobs) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 288.89 Tm (The study doesn't just identify exposed occupations. It also breaks down specific tasks where AI is) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 273.89 Tm (being used.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 251.89 Tm (Inside programming roles, for instance, AI is commonly used to:) Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 46 229.89 Tm (- Write and update software code) Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 46 213.09 Tm (- Generate documentation) Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 46 196.29 Tm (- Assist with debugging or refactoring) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 179.49 Tm (For customer service teams, AI tools help with:) Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 46 157.49 Tm (- Answering routine customer questions) Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 46 140.69 Tm (- Processing orders) Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 46 123.89 Tm (- Responding to complaints or support requests) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 107.09 Tm (In data entry roles, models are used to:) Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 46 85.09 Tm (- Read source documents) Tj ET q 0.86 0.88 0.92 RG 1 w 46 42 m 549.28 42 l S Q BT /F1 8.4 Tf 0.42 0.45 0.5 rg 1 0 0 1 46 30 Tm (TechRounder | Page 2 of 6) Tj ET BT /F1 7.2 Tf 0.42 0.45 0.5 rg 1 0 0 1 46 19 Tm (https://www.techrounder.com/pdf/blog/ai-is-hitting-white-collar-jobs-first-as-new-research-reveals-the-real-automation-frontline.pdf) Tj ET endstream endobj 9 0 obj << /Type /Page /Parent 2 0 R /MediaBox [0 0 595.28 841.89] /Resources << /Font << /F1 3 0 R /F2 4 0 R >> >> /Contents 10 0 R >> endobj 10 0 obj << /Length 4602 >> stream BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 46 789.89 Tm (- Extract structured information) Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 46 773.09 Tm (- Populate digital records automatically) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 756.29 Tm (Medical record specialists are using AI to:) Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 46 734.29 Tm (- Compile patient information) Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 46 717.49 Tm (- Summarize clinical data) Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 46 700.69 Tm (- Code records for administrative systems) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 683.89 Tm (Market research analysts are leaning on AI for tasks like:) Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 46 661.89 Tm (- Preparing written reports) Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 46 645.09 Tm (- Summarizing research findings) Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 46 628.29 Tm (- Generating data explanations and visualizations) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 611.49 Tm (Look closely and a pattern appears.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 589.49 Tm (These tasks all involve transforming information into other forms of information. That's exactly) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 574.49 Tm (where language models shine.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 552.49 Tm (They read text, restructure it, summarize it, classify it, or generate new content from it.) Tj ET BT /F2 15 Tf 0.08 0.1 0.14 rg 1 0 0 1 46 524.49 Tm (So Far, No Major Employment Collapse) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 500.49 Tm (Despite the high exposure levels, the study stops well short of claiming that AI has already triggered) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 485.49 Tm (widespread job losses.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 463.49 Tm (In fact, Anthropic reports no systematic rise in unemployment among workers in highly exposed) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 448.49 Tm (occupations since late 2022, when generative AI began spreading rapidly.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 426.49 Tm (That matters. A lot.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 404.49 Tm (Public conversation around AI often swings between extremes. Either the technology will replace) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 389.49 Tm (everyone immediately, or it will remain a harmless productivity tool.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 367.49 Tm (Reality, at least for now, appears far more gradual.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 345.49 Tm (Employment data simply does not show a sudden collapse in white-collar jobs.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 323.49 Tm (But there are early signals that something subtle may be happening beneath the surface.) Tj ET BT /F2 15 Tf 0.08 0.1 0.14 rg 1 0 0 1 46 295.49 Tm (The First Warning Sign Appears in Hiring) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 271.49 Tm (The clearest early signal in the study appears in entry-level hiring.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 249.49 Tm (Among workers aged 22 to 25, the researchers found that entry into highly exposed occupations) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 234.49 Tm (began to weaken around 2024.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 212.49 Tm (The shift isn't dramatic. But it is noticeable.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 190.49 Tm (Job-finding rates for young workers entering exposed fields dropped by roughly half a percentage) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 175.49 Tm (point, and the post-ChatGPT estimate suggests about a 14 percent decline in job-finding rates compared) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 160.49 Tm (with 2022.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 138.49 Tm (Anthropic is careful not to overstate the result. The statistical signal is weak, and other explanations) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 123.49 Tm (could exist.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 101.49 Tm (Still, the trend lines are worth watching.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 79.49 Tm (Hiring slowdowns tend to show up long before layoffs do.) Tj ET q 0.86 0.88 0.92 RG 1 w 46 42 m 549.28 42 l S Q BT /F1 8.4 Tf 0.42 0.45 0.5 rg 1 0 0 1 46 30 Tm (TechRounder | Page 3 of 6) Tj ET BT /F1 7.2 Tf 0.42 0.45 0.5 rg 1 0 0 1 46 19 Tm (https://www.techrounder.com/pdf/blog/ai-is-hitting-white-collar-jobs-first-as-new-research-reveals-the-real-automation-frontline.pdf) Tj ET endstream endobj 11 0 obj << /Type /Page /Parent 2 0 R /MediaBox [0 0 595.28 841.89] /Resources << /Font << /F1 3 0 R /F2 4 0 R >> >> /Contents 12 0 R >> endobj 12 0 obj << /Length 4491 >> stream BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 789.89 Tm (Companies often adjust by hiring fewer junior employees rather than cutting existing staff.) Tj ET BT /F2 15 Tf 0.08 0.1 0.14 rg 1 0 0 1 46 761.89 Tm (Other Research Is Seeing Something Similar) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 737.89 Tm (This pattern isn't appearing in isolation.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 715.89 Tm (A separate study from the Stanford Digital Economy Lab, using ADP payroll data, found that workers) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 700.89 Tm (aged 22 to 25 in the most AI-exposed occupations saw about a 6 percent decline in employment) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 685.89 Tm (between late 2022 and September 2025.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 663.89 Tm (Older workers in those same occupations, meanwhile, actually saw employment increase.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 641.89 Tm (The study also found the decline was stronger in fields where AI was used for automation rather than) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 626.89 Tm (augmentation.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 604.89 Tm (In simple terms: when AI fully replaces a task instead of assisting with it, employment pressure tends) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 589.89 Tm (to appear sooner.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 567.89 Tm (Anthropic's results are more cautious. But the direction of both studies points to the same possibility.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 545.89 Tm (If AI does reshape the labor market, the earliest impact may fall on newcomers rather than) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 530.89 Tm (experienced professionals.) Tj ET BT /F2 15 Tf 0.08 0.1 0.14 rg 1 0 0 1 46 502.89 Tm (The Demographics of AI Exposure) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 478.89 Tm (Another striking finding from the report is who sits inside the high-exposure group.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 456.89 Tm (Workers in these occupations are typically:) Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 46 434.89 Tm (- More educated) Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 46 418.09 Tm (- Higher paid) Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 46 401.29 Tm (- More likely to hold graduate degrees) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 384.49 Tm (According to the study, employees in the most exposed roles earn about 47 percent more on average) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 369.49 Tm (than workers in the least exposed occupations.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 347.49 Tm (That runs directly against the traditional narrative of automation.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 325.49 Tm (Historically, technology first displaced lower-skill labor - factory work, manufacturing, and routine) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 310.49 Tm (manual jobs.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 288.49 Tm (Generative AI appears to be targeting something different.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 266.49 Tm (It's hitting the structured cognitive layer of the economy - the parts built around documentation, digital) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 251.49 Tm (coordination, research, reporting, coding, and administrative analysis.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 229.49 Tm (These are jobs performed at desks, not on shop floors.) Tj ET BT /F2 15 Tf 0.08 0.1 0.14 rg 1 0 0 1 46 201.49 Tm (Physical Work Remains Largely Outside the Impact Zone) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 177.49 Tm (The study also highlights a group of occupations that currently show no measurable AI exposure at all.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 155.49 Tm (Examples include:) Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 46 133.49 Tm (- Cooks) Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 46 116.69 Tm (- Motorcycle mechanics) Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 46 99.89 Tm (- Lifeguards) Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 46 83.09 Tm (- Bartenders) Tj ET q 0.86 0.88 0.92 RG 1 w 46 42 m 549.28 42 l S Q BT /F1 8.4 Tf 0.42 0.45 0.5 rg 1 0 0 1 46 30 Tm (TechRounder | Page 4 of 6) Tj ET BT /F1 7.2 Tf 0.42 0.45 0.5 rg 1 0 0 1 46 19 Tm (https://www.techrounder.com/pdf/blog/ai-is-hitting-white-collar-jobs-first-as-new-research-reveals-the-real-automation-frontline.pdf) Tj ET endstream endobj 13 0 obj << /Type /Page /Parent 2 0 R /MediaBox [0 0 595.28 841.89] /Resources << /Font << /F1 3 0 R /F2 4 0 R >> >> /Contents 14 0 R >> endobj 14 0 obj << /Length 4695 >> stream BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 46 789.89 Tm (- Dishwashers) Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 46 773.09 Tm (- Dressing room attendants) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 756.29 Tm (These roles require physical presence, spatial awareness, manual dexterity, or direct human) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 741.29 Tm (interaction - capabilities that language models simply do not possess.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 719.29 Tm (That doesn't mean physical work is permanently safe from automation. Robotics continues advancing.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 697.29 Tm (But the tools dominating today's AI wave - large language models - operate almost entirely in the digital) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 682.29 Tm (realm.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 660.29 Tm (And that shapes where the disruption begins.) Tj ET BT /F2 15 Tf 0.08 0.1 0.14 rg 1 0 0 1 46 632.29 Tm (The Broader Labor Market Still Looks Stable) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 608.29 Tm (Stepping back, the overall labor market outlook remains positive.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 586.29 Tm (The U.S. Bureau of Labor Statistics projects that the economy will add about 5.2 million jobs between) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 571.29 Tm (2024 and 2034, representing roughly 3.1 percent employment growth.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 549.29 Tm (Anthropic's study does not suggest AI will reverse that expansion across the entire economy.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 527.29 Tm (However, it does uncover a subtle correlation.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 505.29 Tm (Occupations with higher observed AI exposure tend to have slightly weaker long-term growth) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 490.29 Tm (projections.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 468.29 Tm (For every 10-percentage-point increase in exposure, projected employment growth falls by about 0.6) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 453.29 Tm (percentage points.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 431.29 Tm (That's not definitive proof of displacement.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 409.29 Tm (But it does hint that the jobs where AI is already most useful may not be the strongest growth engines in) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 394.29 Tm (the coming decade.) Tj ET BT /F2 15 Tf 0.08 0.1 0.14 rg 1 0 0 1 46 366.29 Tm (The AI Labor Debate Is Entering a New Phase) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 342.29 Tm (For the past few years, discussions about AI and jobs have often been dominated by extremes.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 320.29 Tm (Some argue that AI will replace huge segments of the workforce almost overnight. Others insist it will) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 305.29 Tm (simply make workers more productive without reducing employment.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 283.29 Tm (The evidence emerging now suggests something more complicated.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 261.29 Tm (The disruption may arrive slowly and unevenly.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 239.29 Tm (Instead of mass layoffs, the early effects could appear as:) Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 46 217.29 Tm (- Slower hiring) Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 46 200.49 Tm (- Fewer entry-level roles) Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 46 183.69 Tm (- Automation of routine tasks) Tj ET BT /F1 10.5 Tf 0.2 0.23 0.28 rg 1 0 0 1 46 166.89 Tm (- Smaller teams doing the same work) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 150.09 Tm (A company might keep the same number of employees but hire fewer junior analysts. A support team) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 135.09 Tm (might handle more tickets without expanding staff. A developer assisted by AI might produce the output) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 120.09 Tm (of several people from a decade ago.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 98.09 Tm (None of those changes look dramatic on their own.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 76.09 Tm (But over time, they reshape professions.) Tj ET q 0.86 0.88 0.92 RG 1 w 46 42 m 549.28 42 l S Q BT /F1 8.4 Tf 0.42 0.45 0.5 rg 1 0 0 1 46 30 Tm (TechRounder | Page 5 of 6) Tj ET BT /F1 7.2 Tf 0.42 0.45 0.5 rg 1 0 0 1 46 19 Tm (https://www.techrounder.com/pdf/blog/ai-is-hitting-white-collar-jobs-first-as-new-research-reveals-the-real-automation-frontline.pdf) Tj ET endstream endobj 15 0 obj << /Type /Page /Parent 2 0 R /MediaBox [0 0 595.28 841.89] /Resources << /Font << /F1 3 0 R /F2 4 0 R >> >> /Contents 16 0 R >> endobj 16 0 obj << /Length 2682 >> stream BT /F2 15 Tf 0.08 0.1 0.14 rg 1 0 0 1 46 789.89 Tm (The Office Is Becoming the First Testing Ground) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 765.89 Tm (The most important takeaway from Anthropic's research isn't that an AI employment crisis has) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 750.89 Tm (already arrived.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 728.89 Tm (It's that the first real pressure points are becoming visible.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 706.89 Tm (The office - not the factory - is emerging as the initial testing ground for AI substitution.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 684.89 Tm (Knowledge work built around documents, analysis, coding, reporting, and digital coordination sits) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 669.89 Tm (directly in the path of language-model capability.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 647.89 Tm (That doesn't mean these jobs disappear.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 625.89 Tm (Many will evolve instead.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 603.89 Tm (But one long-held assumption is becoming harder to defend: the idea that white-collar knowledge work) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 588.89 Tm (is naturally protected from automation.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 566.89 Tm (If generative AI continues improving at its current pace, the first major employment shock may not) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 551.89 Tm (start on an assembly line.) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 529.89 Tm (It may begin quietly in places most people never expected - inside help desks, research reports,) Tj ET BT /F1 11 Tf 0.14 0.16 0.2 rg 1 0 0 1 46 514.89 Tm (spreadsheets, and code editors.) Tj ET BT /F2 13 Tf 0.08 0.1 0.14 rg 1 0 0 1 46 486.89 Tm (References) Tj ET BT /F1 10 Tf 0.18 0.2 0.24 rg 1 0 0 1 46 466.89 Tm (1. anthropic.com - research / labor-market-impacts -) Tj ET BT /F1 10 Tf 0.18 0.2 0.24 rg 1 0 0 1 46 453.39 Tm (https://www.anthropic.com/research/labor-market-impacts) Tj ET BT /F1 10 Tf 0.18 0.2 0.24 rg 1 0 0 1 46 435.89 Tm (2. digitaleconomy.stanford.edu - app / uploads -) Tj ET BT /F1 10 Tf 0.18 0.2 0.24 rg 1 0 0 1 46 422.39 Tm (https://digitaleconomy.stanford.edu/app/uploads/2025/11/CanariesintheCoalMine_Nov25.pdf) Tj ET q 0.86 0.88 0.92 RG 1 w 46 42 m 549.28 42 l S Q BT /F1 8.4 Tf 0.42 0.45 0.5 rg 1 0 0 1 46 30 Tm (TechRounder | Page 6 of 6) Tj ET BT /F1 7.2 Tf 0.42 0.45 0.5 rg 1 0 0 1 46 19 Tm (https://www.techrounder.com/pdf/blog/ai-is-hitting-white-collar-jobs-first-as-new-research-reveals-the-real-automation-frontline.pdf) Tj ET endstream endobj xref 0 17 0000000000 65535 f 0000000015 00000 n 0000000064 00000 n 0000000154 00000 n 0000000224 00000 n 0000000299 00000 n 0000000441 00000 n 0000005778 00000 n 0000005920 00000 n 0000010629 00000 n 0000010772 00000 n 0000015426 00000 n 0000015570 00000 n 0000020113 00000 n 0000020257 00000 n 0000025004 00000 n 0000025148 00000 n trailer << /Size 17 /Root 1 0 R >> startxref 27882 %%EOF