Mercor: 90% of US work hours lack good AI benchmark coverage
Mercor's Labor Index study mapped 29 AI benchmarks onto O*NET's 923 US occupations and 18,797 tasks, finding that only 208 of 923 occupations have a task covered by at least five benchmark prompts, and 6 of 22 job families have no such coverage. It estimates that about 90% of American working hours go to tasks with little or no coverage from existing benchmarks. The report says coverage is not random: tasks tagged as AI-exposed are roughly 6x more likely to appear in benchmarks, and tasks in occupations with larger wage bills or rated most important within their occupation are also more likely to be covered (the latter roughly 6–10x), though the authors stress the results are provisional and will be updated as the benchmark list grows.
Benchmark blind spots shape how enterprises, policymakers and workers judge what models can actually do at work.
The Benchmark Map
Each tile represents one of the 923 ONET occupations making up the US labor force, grouped into 22 job families. Occupations are shaded according to the share of their tasks which are represented in our initial list of AI benchmarks. A single benchmark prompt can be mapped to multiple ONET tasks.
click a family to open it
Coverage by
Covered with at least
- Benchmark Index
- No coverage