The Remote Labor Index (RLI) is a benchmark that measures how well AI agents can complete real freelance work that humans normally perform remotely. It was released in 2025 by the Center for AI Safety (CAIS) and Scale AI. Its tasks are drawn from projects actually posted on freelance platforms across fields such as graphic design, video editing, data analysis, software development and writing, and human experts judge whether the deliverables meet the client's requirements.
The primary metric is the automation rate, the share of projects an agent completes to a professional standard. Unlike benchmarks built around coding puzzles or multiple-choice exams, RLI evaluates multi-step, end-to-end work by its final deliverables, and it is used as an indicator of how much remote labor AI systems can currently take on.