Multi-source labor analysis

The AI Exposure Atlas: where forecasts collide with deployment reality

342 occupations, four independent frameworks, one joined dataset. This page visualizes confrontations between theoretical AI capability, observed usage, and pre-LLM automation baselines.

Download unified CSV View source repository

Scroll to explore source-by-source disagreement

Occupation Explorer

The joined occupation table is the core asset in this repo. Filter it, rank it by any major metric, and inspect how the four source views line up for a single job.

Occupation
Consensus
Workers
Sources

Workforce Exposure Frontier

The most consequential occupations are not just highly exposed, they also employ millions of people. This view surfaces large labor pools where consensus exposure is already elevated.

X = consensus exposure, Y = occupation employment on a log scale, bubble size = median pay, color = disagreement.

Pre-LLM vs LLM Inversion

Frey and Osborne captured a robotics-era view of automation. Karpathy captures LLM-era cognitive exposure. The diagonal is equality; points below it are occupations that looked automatable in 2013 but less exposed to current LLM workflows.

Theory-Reality Gap

OpenAI beta estimates potential task-time reduction. Anthropic observed exposure tracks real deployment. Distance below the diagonal is the adoption ceiling.

Consensus vs Disagreement

Consensus is average score across available LLM-era sources; disagreement is max minus min. High consensus + low disagreement are robust signals. High disagreement flags methodological blindspots.

LLM Disruption Delta Occupations

Occupations where Frey and Osborne rated low risk (`<0.3`) but Karpathy rates high (`>=0.7`). This isolates jobs newly exposed by LLMs.

Category-Level Confrontation Map

Workforce-weighted category averages by source. This highlights where all sources align (or conflict) at sector scale.

Researcher Paradox Dashboard

Developer-class work is both highly exposed and highly rewarded. The same labor segment appears central to automation and to economic demand.