US AI leaders turn to Chinese open-weight models, challenging closed-source safety claims

More American titans of artificial intelligence are describing Chinese open-weight models as better for AI safety and security than closed-source models, challenging the long-standing claim by US closed-source AI model developers such as…