I believe that allowing the proliferation of open-weight models capable of recursive self-improvement can drive remarkable innovation and offer substantial benefits to society. By embracing open development, we foster collaboration and transparency, which are essential for ethical and responsible AI advancements.