PANews reported on September 26th that OpenAI launched a new evaluation tool, GDPval , which focuses on measuring AI performance on real-world economic value tasks. GDPval covers 44 occupations across the nine industries that contribute most to US GDP . The tasks were designed by industry experts with an average of 14 years of experience. Evaluation results show that nearly half of the outputs of the Claude Opus 4.1 model are comparable to or better than expert performance. OpenAI stated that it will continue to expand the scope and details of GDPval's evaluations in the future.


