The 25 Jobs Most at Risk from AI in 2026

Every week brings new headlines about AI replacing workers. But which jobs are actually most at risk? Not the hypothetical, science-fiction version of AI — but the AI tools that exist right now, in production, being used by real companies.

We analyzed data from Anthropic's economic impact research, the World Economic Forum's Future of Jobs Report, McKinsey Global Institute projections, and real-world layoff data from 2024-2026 to rank the 25 most vulnerable occupations.

How We Ranked These Jobs

Each job was scored on four factors:

1. Task Automation Potential (40% weight) — What percentage of daily tasks can current AI tools perform at acceptable quality?
2. Real-World Displacement Evidence (30% weight) — Are companies actually cutting these roles and citing AI?
3. Speed of Adoption (20% weight) — How quickly is the industry adopting AI for this function?
4. Recovery Difficulty (10% weight) — How hard is it for displaced workers to transition to similar roles?

Scores range from 0-100, with 100 being the highest risk.

The Rankings

Tier 1: Critical Risk (Score 85-100)

1. Data Entry Clerk — Risk Score: 97

AI handles structured data input faster, cheaper, and with fewer errors than humans. OCR + LLMs can process forms, invoices, and documents with 99%+ accuracy. Major companies have already automated 90%+ of data entry.

Jobs at risk globally: 8.2 million Median U.S. salary being displaced: $37,000 Timeline: Already happening — 60% reduction by 2028

2. Telemarketer — Risk Score: 95

AI voice agents can make outbound calls, handle objections, and qualify leads at scale. Companies like Bland.ai and Vapi are replacing entire call center floors.

Jobs at risk globally: 4.8 million Median U.S. salary being displaced: $31,000 Timeline: 70% reduction by 2028

3. Bookkeeper / Accounting Clerk — Risk Score: 93

AI accounting tools (QuickBooks AI, Xero AI, custom LLM integrations) automate transaction categorization, reconciliation, and basic financial reporting.

Jobs at risk globally: 6.1 million Median U.S. salary being displaced: $45,000 Timeline: 50% reduction by 2029

4. Customer Service Representative (Tier 1) — Risk Score: 91

AI chatbots and voice agents now resolve 80%+ of standard customer queries. Anthropic, OpenAI, and Google all offer enterprise customer service AI products.

Jobs at risk globally: 12.5 million Median U.S. salary being displaced: $38,000 Timeline: 60% reduction by 2028

5. Proofreader / Copy Editor — Risk Score: 89

LLMs now match or exceed human proofreaders for grammar, style, and consistency checks. Combined with domain-specific fine-tuning, AI editing tools are production-ready.

Jobs at risk globally: 1.2 million Median U.S. salary being displaced: $44,000 Timeline: 50% reduction by 2028

Tier 2: High Risk (Score 70-84)

6. Paralegal / Legal Assistant — Risk Score: 84

AI legal research tools (Harvey, CoCounsel) can review contracts, summarize case law, and draft legal documents. Baker McKenzie already cut 800 positions citing AI.

Jobs at risk globally: 3.4 million Timeline: 40% reduction by 2029

7. Basic Financial Analyst — Risk Score: 82

AI models can analyze financial statements, build models, and generate investment reports. Junior analyst roles at Goldman Sachs, BlackRock, and JPMorgan are being reduced.

Jobs at risk globally: 2.8 million Timeline: 35% reduction by 2029

8. Medical Transcriptionist — Risk Score: 81

AI speech-to-text with medical vocabulary has reached 97%+ accuracy. Most hospitals and clinics are transitioning to automated transcription.

Jobs at risk globally: 1.1 million Timeline: 60% reduction by 2028

9. Travel Agent — Risk Score: 79

AI travel planning tools combine real-time pricing, preference learning, and itinerary optimization. The remaining value is in luxury/complex travel where human judgment adds premium value.

Jobs at risk globally: 2.1 million Timeline: 40% reduction by 2029

10. Insurance Underwriter — Risk Score: 77

AI models assess risk, price policies, and process applications faster and more consistently than human underwriters for standard cases. Complex commercial underwriting remains human-dominant.

Jobs at risk globally: 1.5 million Timeline: 35% reduction by 2030

11. Tax Preparer — Risk Score: 76

AI tax software handles standard returns with high accuracy. The IRS Direct File program and AI-powered tools like TurboTax AI are reducing demand for human preparers.

Jobs at risk globally: 1.8 million Timeline: 40% reduction by 2029

12. Quality Assurance Tester (Manual) — Risk Score: 75

AI testing tools generate test cases, run automated regression tests, and identify bugs. Manual QA is rapidly being automated across the software industry.

Jobs at risk globally: 3.2 million Timeline: 45% reduction by 2029

13. Technical Writer — Risk Score: 73

AI generates documentation from code, API specs, and product descriptions. Oracle, Microsoft, and Google have all reduced technical writing teams.

Jobs at risk globally: 1.4 million Timeline: 35% reduction by 2029

14. Translator (Non-Literary) — Risk Score: 72

AI translation quality for business and technical documents now approaches human quality for major language pairs. Literary and creative translation remains human-dominant.

Jobs at risk globally: 2.6 million Timeline: 40% reduction by 2030

15. Receptionist — Risk Score: 71

AI phone systems and digital reception tools handle scheduling, routing, and basic inquiries. Physical front-desk presence remains valued in some industries.

Jobs at risk globally: 4.1 million Timeline: 35% reduction by 2029

Tier 3: Elevated Risk (Score 55-69)

16. Real Estate Appraiser — Risk Score: 68

AI models analyze comparable sales, market trends, and property features. Still requires physical inspection for complex properties.

17. Loan Officer — Risk Score: 66

AI credit scoring and loan processing automation. Human judgment still needed for non-standard applications.

18. Market Research Analyst — Risk Score: 65

AI tools scrape, analyze, and synthesize market data faster. Human insight still valued for strategic interpretation.

19. Recruiter (Agency) — Risk Score: 63

AI screening, matching, and initial outreach automation. Relationship-based recruiting less affected.

20. Graphic Designer (Production) — Risk Score: 61

AI image generation handles templates, social media graphics, and routine design. Creative direction and brand strategy remain human.

21. Journalist (Beat Reporter) — Risk Score: 59

AI generates routine news stories (earnings, sports scores, weather). Investigative and feature journalism remains human.

22. Radiologist — Risk Score: 58

AI image analysis matches or exceeds human radiologists for many scan types. Regulatory and liability frameworks slow adoption.

23. Truck Driver (Long-Haul) — Risk Score: 57

Autonomous trucking technology is production-ready for highway segments. Last-mile and complex urban driving still requires humans.

24. Warehouse Worker — Risk Score: 56

Amazon and others deploying robots at scale. Physical adaptability in unstructured environments still limits full automation.

25. Retail Cashier — Risk Score: 55

Self-checkout and cashierless stores expanding. Customer preference and theft prevention keep some human cashiers.

What the Data Tells Us

Pattern 1: Routine Cognitive Work Is Most Exposed

The highest-risk jobs share a common trait: they involve routine cognitive tasks — processing information, following established rules, generating standard outputs. Physical jobs are less exposed because robotics lags behind AI software.

Pattern 2: Junior/Entry-Level Roles Are Hit First

AI tends to automate the tasks that junior employees do — data processing, basic analysis, standard document creation. This creates a "missing rung" problem: how do you become a senior expert if the entry-level pathway is automated?

Pattern 3: The "Last Mile" Protects Some Jobs

Jobs that require physical presence in unpredictable environments (plumbers, electricians, nurses) have natural protection. The cost of robotics that can handle these environments is still prohibitive.

Pattern 4: Regulation Slows (But Does Not Stop) Displacement

Healthcare, legal, and financial services have regulatory barriers that slow AI adoption. But these are speed bumps, not roadblocks. Expect 3-5 year delays compared to unregulated industries.

Is Your Job on This List?

If your role appears above, do not panic — but do act. Here is what to do:

1. Get a personalized assessment. General rankings are useful, but your specific situation matters. Take the LayoffReady Risk Assessment for a score based on YOUR industry, company, and role.

2. Identify the un-automatable parts of your work. Every role has tasks that AI cannot do well — complex judgment, creative thinking, relationship building, physical dexterity. Double down on these.

3. Learn to work with AI, not against it. The goal is not to compete with AI but to become more valuable by using it. An accountant who uses AI tools is 5x more productive than one who does not.

4. Build financial resilience. If your role is in Tier 1 or 2, aim for 6-12 months of expenses in savings. Career transitions take time.

5. Start networking now. 80% of jobs come through connections. Build your network while you are employed, not after you are laid off.

The Jobs AI Is Creating

For balance, here are the fastest-growing roles that AI is creating:

• AI/ML Engineer (+47% YoY demand)
• Prompt Engineer / AI Product Manager (+120% YoY)
• AI Safety / Ethics Specialist (+85% YoY)
• Data Quality Manager (+35% YoY)
• AI Integration Specialist (+65% YoY)
• Robotics Maintenance Technician (+28% YoY)
• AI Training Data Specialist (+55% YoY)

The net job impact is negative in the short term, but the new roles tend to pay significantly more than the ones being eliminated.

---

Where does your job rank? Take the LayoffReady Risk Assessment for a personalized risk score based on your specific role, industry, and company. It takes 2 minutes and has been used by over 47,000 workers.