Today we're looking at the newest OpenAI publication introducing GDPval, a new evaluation designed to measure the performance of AI models on economically valuable, real-world tasks. Source: https://openai.com/index/gdpval/The paper: https://cdn.openai.com/pdf/d5eb7428-c4e9-4a33-bd86-86dd4bcf12ce/GDPval.pdfThis evaluation spans 44 knowledge work occupations across nine major industries contributing to U.S. GDP, moving beyond traditional academic benchmarks to focus on realistic work products like legal briefs and engineering blueprints. Tasks are meticulously developed and graded by experienced industry professionals, who compare outputs from leading models, such as GPT-5 and Claude Opus 4.1, against human-produced work. Early results indicate that frontier models are rapidly approaching expert quality in many areas, performing tasks significantly faster and cheaper, though the evaluation currently has limitations, such as not capturing complex, multi-draft workflows. OpenAI aims to use GDPval to transparently track AI progress and understand its potential impact on the future of work.#openai #artificialintelligence #ai #gdpval #economy ___What do you think?PS, make sure to follow my:Main channel: https://www.youtube.com/@swetlanaAIMusic channel: https://www.youtube.com/@Swetlana-AI-Music Hosted on Acast. See acast.com/privacy for more information.

Ver todos los episodios