Will AI
take your job?

AI rarely eliminates an entire occupation overnight. Instead, AI automates or accelerates specific tasks within a job. Roles dominated by routine, screen-based data processing experience high workflow change, while jobs requiring physical dexterity, unconstrained environment navigation, deep stakeholder trust, and legal accountability remain anchored by human professionals.

AI Exposure (0–100) measures how much of an occupation's day-to-day task mix overlaps with the technical capabilities of current artificial intelligence systems. A high exposure score means software can perform or assist with a significant share of the workload.

Replacement Risk (0–100) estimates whether technical AI exposure translates into reduced human demand. It accounts for real-world friction—including physical constraints, human dependency, professional accountability, adoption economics, and institutional regulations.

No. JobsVsAI scores are index ratings on a 0–100 scale, not probabilities or unemployment forecasts. A score of 70 indicates that the occupation exhibits higher structural vulnerability compared to lower-scoring occupations across the economy.

A job can be highly exposed to AI without being easy to replace. For example, software developers or healthcare practitioners perform many analytical and administrative tasks AI can accelerate, yet human architecture, regulatory liability, and bedside care prevent direct headcount reduction.

JobsVsAI uses official occupational taxonomy and task data from the U.S. Department of Labor's O*NET database (Version 30.3), encompassing detailed task ratings, work activities, and importance weights.

We evaluate every task in an occupation against 15 AI capability dimensions from our Capability Index, compute task-level automation feasibility, aggregate weighted task exposure, and then apply structural modifiers for physical work, human dependency, and commercial adoption pressure.

No chatbot generates the score when a page is opened. JobsVsAI scores are calculated through a versioned scoring pipeline from occupational task evidence and mapped AI capabilities.

Scores are recalculated when new major AI capability milestones are verified, when underlying O*NET taxonomy releases occur, or when methodology calibrations are published in our changelog.

Yes. Local adoption speed, wages, regulatory requirements, and union protections alter how quickly businesses invest in automation. JobsVsAI scores reflect baseline occupation structure rather than local market variations.

High risk is a signal to adapt, not panic. Workers in exposed fields should proactively adopt AI tools to accelerate routine tasks while leaning into interpersonal leadership, specialized domain judgment, and complex verification work that algorithms cannot replace.