Roles overview
How much of a role's task mix overlaps with what current AI can draft, aggregated across people who chose to share an anonymous summary of their reading. Each row needs at least 5 readings before it appears, so no row describes one person.
Read it as augmentation, not replacement. An “exposed” task is one where AI can produce a first draft today; a person still reviews, corrects and owns the result. The score is the average share of a role's tasks that are exposed or worth watching, with every task weighing the same.
Contributions are opt-in and anonymous: role bucket, industry bucket, seniority band, score, level counts and the month. No task text, no identifiers. Per the terms, this page must not be used for recruitment, selection, evaluation or any other decision about an individual.