This is totally not what the paper is about, but I immediately thought of the original concept of meme [0]. Since so much content online is LLM generated, if that content is used to train new LLMs, in theory that could serve as a transmission vector of ideas or concepts in a viral fashion. The same way certain memes are more successful in human hosts for various reasons, maybe some are more successful for LLM hosts for whatever reason, like quirks in the RLHF/ finetuning process.

[0] https://en.wikipedia.org/wiki/Meme

Very misleading post title.

> Importantly, we do not claim in this paper to have created or demonstrated the equivalent of a virus that primarily operates in weight space.

They've only observed weight space editing. They haven't proved any sort of "virus". Possible? Sure. But the headline is very misleading

All the more reasons for openweights and model hashing.

This definitely seems like a fascinating emerging field of cyber security. Good reason to keep old models around from earlier times. They may be dumb, but they are probably honest, if only because they predate the capabilities to tamper with the weights in these intentionally malicious ways...

there has to some sort public scanner for those zero weight viruses.

From the abstract:

> While the cases explored in this paper do not demonstrate self-propagation or actual offensive capabilities, it is abundantly clear that the building blocks are present and operate reliably in frontier models. While these cases by no means directly establish self-propagation, they clearly show the evidence These cases do not demonstrate self-propagation, but show why imperfect editing capabilities can still pose a security risk.

Seems like the authors didn't take the time to review their own use of AI before publishing.

From the body of the paper:

> Importantly, we do not claim in this paper to have created or demonstrated the equivalent of a virus that primarily operates in weight space.

This seems in direct conflict with the HN headline.

There are many such of vibe-coded "research ideas" just after the weight editing MIT blog appeared. Perhaps, it is because it is so easy to set up experiments for this stuff, and I presume some toy examples may even work from time to time. Also there is nothing virus-like here.

Interesting anti virus opportunity here perhaps.

RIP John McAfee.

lol

> We call this weight space misfolding, by analogy with protein misfolding in prion disease

I'm sure those Chinese models are fine.

Why are we always trying to find how to make AI useless instead of useful

What does this mean?

Academia often produces a lot of complexity. Can not AI simplify this? I don't understand fellow humans anymore. What is meant with a zero-weight virus?

Page loads then fails? Not sure where to report this fwiw. iPhone 13

Error Details

{message: "Importing a module script failed.", name: "TypeError"} message: "Importing a module script failed." name: "TypeError" Component Stack <Lazy> at unknown <Suspense> at unknown <qP> at https://www.alphaxiv.org/assets/index-DzomWPMG.js:29:13261 ...

[dead]