The 10 Jobs Most Likely to Be Replaced by AI
We scored every occupation in the U.S. economy across six risk dimensions using data from 12 research institutions. These are the jobs that landed at the top — and the pattern tells a bigger story than any single headline.
Forget the vague predictions. We built a scoring model that analyzes 997 occupations across six measurable dimensions — task automation potential, cognitive AI exposure, physical requirements, creativity demand, social intelligence, and regulatory barriers — using peer-reviewed research from Oxford, Stanford, McKinsey, the OECD, and eight other institutions.
Every job gets a composite score from 1 to 10. Higher means more vulnerable. The results aren't random — they reveal a clear pattern about which kinds of work AI targets first, and why some jobs that seem safe actually aren't.
The Pattern You'll Notice
The top 10 isn't dominated by factory workers or truck drivers — it's almost entirely white-collar, finance, and administrative roles. Jobs built on processing information, following rules, and making decisions from structured data. That's exactly what large language models and AI agents are optimized to do.
The Top 10: Highest AI Displacement Risk
| # | Occupation | Risk Score | Category |
|---|---|---|---|
| 1 | Accountants & Auditors | 9.5 | Finance & Accounting |
| 2 | Budget Analysts | 9.5 | Finance & Accounting |
| 3 | Loan Officers | 9.5 | Finance & Accounting |
| 4 | Credit Clerks & Checkers | 9.5 | Administrative |
| 5 | Credit Counselors | 9.4 | Finance & Accounting |
| 6 | Bookkeepers & Auditing Clerks | 9.4 | Administrative |
| 7 | Billing & Posting Clerks | 9.2 | Administrative |
| 8 | Payroll & Timekeeping Clerks | 9.2 | Administrative |
| 9 | Receptionists & Information Clerks | 9.2 | Administrative |
| 10 | Switchboard Operators | 9.1 | Administrative |
Why These Jobs, Specifically?
Financial analysis, tax prep, and audit procedures are increasingly automated by AI systems trained on regulatory frameworks.
Budget forecasting, variance analysis, and fiscal reporting follow structured patterns that AI handles with near-human accuracy.
Credit assessment, risk scoring, and loan approval workflows are already heavily algorithm-driven at major lenders.
Evaluating creditworthiness and processing applications is a pattern-recognition task that AI excels at.
Debt analysis and repayment planning are becoming commoditized through AI-powered financial advisory tools.
Transaction categorization, reconciliation, and ledger management are among the most automatable tasks in the economy.
Invoice processing, payment posting, and billing dispute resolution are high-volume repetitive processes.
Payroll calculation, tax withholding, and time tracking follow strict rules — exactly the kind of logic AI handles best.
AI chatbots and virtual assistants are already replacing first-contact roles in healthcare, legal, and corporate settings.
Call routing, message taking, and basic inquiry handling are mature automation targets with existing deployments.
How We Calculate These Scores
Every occupation is analyzed across six independent dimensions. Three measure vulnerability (how exposed the job is to automation), and three measure resilience (structural factors that protect it). The composite score balances all six.
Task Automation Potential
How many of the job's core tasks can current or near-term AI systems perform? Based on O*NET task descriptions mapped against AI capability benchmarks.
Cognitive AI Exposure
How exposed are the job's cognitive requirements to AI substitution? Routine analytical work scores high; novel problem-solving scores low.
Physical Requirement
Jobs requiring physical presence and manual dexterity are harder to automate. Software AI moves faster than robotics — low physical demand means higher risk.
Creativity Demand
Original thinking, artistic expression, and cultural context remain resistant to AI. High creativity demand acts as a protective shield.
Social Intelligence
Deep human connection, emotional intelligence, and relationship management are among the last frontiers for AI to crack.
Regulatory Barriers
Licensing requirements, legal mandates, and professional certifications create structural barriers that slow displacement regardless of technical capability.
Composite Formula
Vulnerability − Resilience = AI Risk Score (1–10)
Task Automation + Cognitive Exposure + Physical Vulnerability, weighted against Creativity + Social Intelligence + Regulatory Barriers
The Data Behind the Numbers
This isn't one model or one opinion. Our scoring system synthesizes findings from 12 research institutions spanning government labor statistics, academic peer-reviewed studies, international economic organizations, and AI capability research labs.
997
Occupations Scored
12
Research Institutions
6
Risk Dimensions
Our Data Sources
What This Means for You
If your job is on this list, it doesn't mean you'll be unemployed tomorrow. It means the structure of your work is changing — probably faster than you think. The tasks you spend most of your day on are becoming automatable. That gives you a window to adapt, but the window is closing.
If your job isn't on this list, don't assume you're safe. A 5.0 score still means significant transformation ahead. And these scores update as AI capabilities evolve.
Either way, the first step is the same: know your actual risk score, understand which of your specific tasks are vulnerable, and build a plan before the market forces you into one.
Where Does Your Job Rank?
Search from 997 scored occupations and get your personalized AI risk score instantly — with a full report including task-by-task analysis, career pivot plan, and skills roadmap.
Check Your Risk Score →Free score preview. Full report $29.99.