Post by @jonny@neuromatch.social

2026-09-29 06:17 UTC

But how would i get such an infected binary on my muse VM? Well! can i interest you in a little "literally the whole thing is designed for that to happen?" The entire idea of the product is that you can ask the thing to go "musey, can you set me up a photo album website and put on my line?" and it's supposed to be able to run some service and host your photos and do whatever. If you prompt muse to do things that would lead it to install software, it will. if you prompt it to do things that would lead it to write software it will. So is there is any affected package from any supply chain attack or typosquat, muse goes npm install or any of the other numerous mechanisms by which installation turns into a vector for arbitrary code execution, then you have meemaw and pawpaw's request to make a photo album installing a token miner without them ever knowing (except for maybe having to bump their subscription tier, token counting is also broken - the number of tokens being reported as used in the billing window is roughly double what gets reported from the token counts in the raw inference api, so, either a) fraud, or b) all the extra machinery bolted on like sentinel roughly doubles token use, fun!)

Replies (1)

@jonny@neuromatch.social 2026-09-29 06:35

That might still sound sort of weak - oh hum but how would you scale typosquatting and supply chain compromise into a botnet from commonly requested actions, aside from the usual ways? This is where the agent spawning, "ideas", and "spaces" come in. Recall the idea of "fleet learning" - https://neuromatch.social/@jonny/117340290530430198 - On September 8th, on this random podcast i guess, (at timestamp 25:27) zuck talks about "fleet learning," and describes it as being a vector for sharing "ideas" between muse agents, which he presents as these little bite-size events like "teach my daughter about history through civilization" (again, as always, "automate the love for my children" takes center stage). This shows up in the "ideas" tab right now. However this concept of "ideas" as a sharable unit of anything is more general, as i describe in quoted post from extracted strings - fleet swapping of "ideas" happens for things that aren't user visible. Also in the extracted binary strings is this yet-to-be-released idea of "spaces." (i'm going to drop the quotes and just capitalize, so just note i'm talking about the meta-brand Ideas and meta-brand Spaces when they are capitalized). in short: Ideas are communicable skill-like prompts that can induce spaces, and spaces are intended to be small, communicable mini-apps that are a unit of some prompt + code that run on your muse vm instance

Open ##4893728