AI Automation Reaches 16.1% in Projects, Impacting Remote Workers
The Remote Labor Index (RLI) shows AI automation rates have surged, with the Claude Fable 5 model achieving 16.1% project completion. This marks a significant leap in AI's capabilities.
The Center for AI Safety (CAIS) has updated its Remote Labor Index (RLI), revealing a rapid increase in AI automation capabilities. The Claude Fable 5 model has reached an automation rate of 16.1% for completing projects, setting a new benchmark in this testing.
The RLI measures the ability of AI agents to perform real-world remote freelance projects according to industrial standards. The tests cover diverse fields including 3D & CAD, architectural design, graphic design, video animation, audio production, data analysis, and web applications. Each deliverable is reviewed by human evaluators against standards set by human professionals.
This 16.1% figure represents a substantial increase from previous benchmarks. Before this update, the highest score recorded was 4.17% by Opus 4.6. Within just eight months of the RLI's launch, the automation rate has grown exponentially.
Specific project examples highlight improvements. Fable 5's ring designs, for instance, showed a notable improvement in quality over prior AI models, though a gap still exists compared to human-designed products. Enhanced visual quality and model accuracy were also observed in 2D advertisement animations and architectural blueprints.