GDPval: Evaluating AI Model Performance on Real-World Economically Valuable Tasks

A preprint introducing GDPval, an evaluation of AI performance on real-world, economically valuable tasks. Both papers cite it in Fig. 3b for the developer-reported share of single tasks done to a professional standard, plotted beside the independent Remote Labor Index. It tracks the capacity to earn that a self-financing lineage would need.

Cited in

both papers (Flagship: Fig. 3; Companion Fig. 3)

Used for

Capability trends; earning; Fig. 3b

Strand

Frontier AI capabilities · Capability trends & benchmarks

Cite

@misc{patwardhan2025gdpval,
  title = {GDPval: Evaluating AI Model Performance on Real-World Economically Valuable Tasks},
  author = {Tejal Patwardhan and Rachel Dias and Elizabeth Proehl and Grace Kim and Michele Wang and Olivia Watkins and Simón Posada Fishman and Marwan Aljubeh and Phoebe Thacker and Laurance Fauconnet and et al.},
  year = {2025},
  url = {https://arxiv.org/abs/2510.04374},
}