自适应和变异的语言模型（LLM）病毒/蠕虫
1 分•作者: rozumbrada•2 天前
我正在思考恶意软件和网络蠕虫的未来。我敢打赌，它们会自我变异，并利用本地模型适应当地环境（一旦这些模型内置于所有设备，并在未来几年内性能足够）。基本上，它们会非常像真正的生物病毒，几乎是真正的有机体。在这种情况下，大型语言模型的非确定性就成了一个特性。每一次感染都可能走上自己独特的发展道路——一半可能夭折，一半可能存活。这就像是基因编程，但却是自主的，并且得到了极大的增强。对于一些非技术（甚至技术？）人士来说，这会让他们联想到“天网”，令人着迷的是，我们正朝着这个方向发展，这突然变得可以想象，理论上很快就能实现。
为什么现在还没有发生？推理成本仍然很高，本地模型也尚未成熟，因此大规模制造它们没有投资回报。但一旦推理成本像电力或自来水一样低廉且本地化，这将是自然而然的发展。届时，我们又该如何阻止其传播呢？
是否已经有一些已记录的实验？
查看原文
I am thinking about a future of malware and cyber worms. I bet it's gonna be self-mutating and adapting to local environment using local models (once they are built-in to all devices and performant enough in future years). Basically almost a real organism resembling real biological viruses. In this case the non-determinism of LLMs is a feature. Every infection could take its own development path - and half might die, half might survive. Think genetic programming but autonomous and on steroids. For some non tech (even tech?) people this reminds Skynet and it's fascinating that we are in a trajectory that this suddenly imaginable and theoretically soon possible.<p>Why is not happening now? Inference is still expensive and local models are not there yet, so there's no ROI in making this at scale. But once inference is local and cheap as electricity or running water, this is the natural development. How do we stop the spreading then?<p>Are there already some documented experiments?