OpenAI says GPT-5 stacks up to humans in a wide range of jobs
-
最新报道：OpenAI introduced GDPval, a new benchmark measuring AI performance against human professionals across 44 occupations in nine key industries. The test shows GPT-5-high matching or beating experts 40.6% of the time, while Anthropic's Claude Opus 4.1 achieved 49% - though OpenAI attributes Claude's high score to its graphic capabilities. While acknowledging current limitations, OpenAI's chief economist notes these results indicate AI's potential to handle routine tasks, allowing professionals to focus on higher-value work. The benchmark represents progress from GPT-4o's 13.7% score 15 months ago.
下一篇
Blue Ocean Technologies joins Pyth Network, bringing US stock nighttime trading data to the blockchain
- According to Bijie.com and ME News, on September 26 (UTC+8), Pyth Network announced that B...
24小时热点
热点专题
413635