Nathan Lambert：开放模型需扩散防御能力
The South Korean bank hack fits with this. Yes, open models are being fitted to be used by some attackers, but so long as Claude Code can be the orchestrator of
AI 导读
韩国银行被黑事件与此吻合。是的，开放模型正被某些攻击者改装利用，但只要 Claude Code 能成为攻击的编排者，我们就迫切需要开放模型来扩散防御能力。这对社会而言是难以承受的艰难平衡，但确有必要。
查看原帖In case you missed my hot take yesterday. https://t.co/Qoopm2J7YB
正文
当前显示原文，中文全文译文尚未就绪。
The South Korean bank hack fits with this. Yes, open models are being fitted to be used by some attackers, but so long as Claude Code can be the orchestrator of a hack we desperately need open models to diffuse defense as well. A hard balance societally to stomach, but needed.