US Accuses Chinese AI of Stealing Technology, Experts Call Claims 'Political' and 'Reckless'
L'essentiel
- US officials have escalated accusations that Chinese AI labs are "stealing proprietary US technology" through distillation, a common industry practice.
- However, AI experts, including Moonshot AI's Randy Xian, have dismissed these claims as "political" and "reckless," arguing that an AI model's output is not copyrighted.
Résumé généré par IA
Pourquoi c'est important
US officials and AI labs have long attributed Chinese AI's rapid advancement to distillation, a common industry practice of using the outputs of larger models to train smaller ones.
In separate posts on X on Thursday, a number of AI experts called the accusations “political” and “reckless”, pointing out that an AI model’s output was not copyrighted.
“Yes, Fable went public on July 1 and K3 launched on July 15. We trained a brand new frontier model in JUST 15 DAYS. Guinness World Record stuff,” Moonshot AI employee Randy Xian wrote on X.
While US officials and AI labs have long attributed Chinese AI’s rapid advancement to distillation, a common industry practice of using the outputs of larger, more capable models to train smaller ones, the White House adviser on Wednesday escalated the accusations by describing the practice as “stealing proprietary US technology”.
Questions ouvertes
- What specific evidence supports the US claims?
- What will be the official response from China?
