Year: 2025
Type: preprint
Abstract: AIs have made rapid progress on research-oriented benchmarks of knowledge and reasoning, but it remains unclear how these gains translate into economic value and automation. To measure this, we introduce the Remote Labor Index (RLI), a broadly multi-sector benchmark comprising real-world, economically valuable projects designed to evaluate end-to-end agent performance in practical settings. AI agents perform near the floor on RLI, with the highest-performing agent achieving an automation rate of 2.5%. These results help ground discussions of AI automation in empirical evidence, setting a common basis for tracking AI impacts and enabling stakeholders to proactively navigate AI-driven labor automation.
Source: arXiv (Cornell University)
Authors: Mazeika, Mantas, Gatti, Alice, Menghini, Cristina, Sehwag, Udari Madhushani, Singhal, Shivam, Orlovskiy, Yury, Basart, Steven, Sharma, Manasi, Peskoff, Denis, Lau, Elaine, Lim, Jaehyuk, Carroll, Lachlan, Blair, Alice, Sivakumar, Vinaya, Basu, Sumana, Kenstler, Brad, Ma, Yuntao, Michael, Julian, Li, Xiaoke, Ingebretsen, Oliver, Mehta, Aditya, Mottola, Jean, Teichmann, John, Yu, Kevin, Shaik, Zaina, Khoja, Adam, Ren, Richard, Hausenloy, Jason, Phan, Long, Htet, Ye, Aich, Ankit, Rabbani, Tahseen, Shah, Vivswan, Novykov, Andriy, Binder, Felix, Chugunov, Kirill, Ramirez, Luis, Geralnik, Matias, Mesura, Hernán, Lee, Dean, Cardona, Ed-Yeremai Hernandez, Diamond, Annette, Yue, Summer, Wang, Alexandr, Liu, Bing, Hernandez, Ernesto, Hendrycks, Dan
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
Funder: Innovationsfonden
Award: G8030081954
OpenAlex ID: W7103153711