中国开源模型威胁美国AI

The entire debate currently taking place is the best PR Chinese open-source labs could hope for. It is not only being openly acknowledged that Chinese models have fewer guardrails and are therefore more useful, but also that they are cheaper and thus pose a threat to the US economy. In doing so, people are indirectly admitting that they see them as serious competition.  > **引用原帖 Chubby♨️ (@kimmonismus):** > China’s open-weight models are starting to threaten the business model behind OpenAI and Anthropic. > Via Wall Street Journal > Kimi K3 and Qwen 3.8 Max are cheap, capable and customizable. Kimi was competitive with leading US systems on some benchmarks, while American companies are increasingly adopting Chinese models to cut AI costs. > That puts pressure on the economics used to justify billions in frontier-model investment and trillions in future infrastructure spending. > OpenAI and Anthropic executives are warning about cybersecurity and national-security risks. Critics point out that both companies are preparing for possible public listings and would benefit from restrictions on cheaper competitors. > Probably explains why the push so hard against chinese open source. > https://x.com/kimmonismus/status/2079518612025430036