TechCrunch旗下Equity播客采访了Hugging Face首席政策官兼战略高级副总裁Irene Solaiman,双方讨论了企业为何转向开源AI模型以降低成本,以及在中国开源模型崛起的背景下美国面临的竞争压力1。Solaiman指出,初创企业和政府正利用开源AI模型以降低成本、增加控制权及灵活性1。然而,这引发了关于商业价值的质疑:当模型免费时,开发者如何捕获价值?1
针对地缘政治观点,Solaiman认为,美国可能需要依靠Nvidia等资金雄厚的参与者构建具有竞争力的开源前沿模型,以与中国竞争1。此外,讨论还聚焦于安全议题,近期对开源AI平台的攻击引发了如何在封闭与开放之间平衡安全的难题1。
Hugging Face’s Chief Policy Officer and Senior Vice President of Strategy, Irene Solaiman, discussed the strategic trade-offs involved in building on open-source artificial intelligence models during an interview on TechCrunch’s Equity podcast. 1 She highlighted that startups and government entities are increasingly turning to open-source AI to reduce costs while gaining greater control and flexibility over their technological infrastructure. 1 However, Solaiman raised critical questions regarding the commercial viability of this approach, specifically asking how developers can capture value when the underlying models are distributed for free. 1
The discussion also addressed the geopolitical implications of the rising prominence of Chinese open-source models. 1 Solaiman suggested that the United States might need to rely on well-capitalized players, such as Nvidia, to develop competitive open-source frontier models in order to maintain its standing against China. 1 Furthermore, she pointed to challenges within the ecosystem, noting that securing open-source platforms remains difficult due to the inherent tension between safety and openness. 1
评论
还没有评论,欢迎留下第一条。