Benchmark

Remote Labor Index (RLI)

Measures: An AI agent must deliver 240 real freelance projects end to end (game development, product design, architecture, data analysis, video animation and more; 23 domains, over 6,000 hours and $140,000 of human work); trained experts manually judge whether a reasonable client would accept each deliverable as at least as good as the human one, and the automation rate is the share that pass.

Maintainer
Scale AI and Center for AI Safety (CAIS)
Introduced
2025
Status
Active
Saturation
Headroom
Typical provenance
Independent
In scoring set
Yes

PaperOfficial leaderboard

Leaderboard

Independent and third-party results only.

No independent or third-party results yet. Show all results