Hacker Newsnew | past | comments | ask | show | jobs | submitlogin
Chinese AI models have lagged the US frontier by 7 months on average since 2023 (epoch.ai)
4 points by stared 30 days ago | hide | past | favorite | 3 comments


"This gap closely resembles the broader gap between proprietary and open-weight models. This is unsurprising since nearly all leading Chinese models are open-weight, while frontier US models remain closed."

Open weights has nothing to do with why the U.S. is leading. That would imply the US is taking advantage to maintain their lead using at the very least, knowledge extracted from competitors open weights.


Do Chinese have closed weight models?

Baidu has https://ernie.baidu.com/ open weight model but do they use closed model in some of their projects or internally?


7 months doesn't change anything. It's not exactly, or otherwise, a strategic moat. Unless you can take over the world in half a year, does it matter?




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: