Post by @Architeuthis@awful.systems

2026-10-13 11:37 UTC

Anthropic has a new angle for trying to get local LLMs banned, apparently they’re about to be able to use your temu android tv to take over your smart toaster: Our experiments suggest that small AI models may be months away from being capable enough to self-replicate on commercial internet-of-things (IoT) devices, like smart ovens and off-brand Android TV boxes. A replicator like this would be challenging to shut down, since there’s no centralized point to shut off—though it may be slow (with a doubling time of days). Unlike non-AI malware, it could also figure out unfamiliar devices on its own. In our evaluations, a 9B model got inference running on 8 of the 9 hardware types we tested

Replies (7)

@samvines@awful.systems 2026-10-13 12:23

I really can’t decide whether those schmucks at Anthropic are deliberately playing up the capabilities of next token prediction machines for the PR/Regulatory capture angle or if they really are earnestly so completely in the thralls of AI psychosis that they actually believe their own shit. I’m fairly sure sama and the gang fall into camp A but with Dario’s not-so-merry-men I’m not so sure…

Open ##5013246
@fullsquare@awful.systems 2026-10-13 13:21

rats’ best thought out attempt at immanentizing eschaton

Open ##5013608
@lagrangeinterpolator@awful.systems 2026-10-13 15:21

It’s funny how they’re twisting themselves into pretzels to make this work. They started by casually saying that AI could copy itself onto the internet, not anticipating that people would immediately point out just how much hardware it takes to run their AI. Seems like they really want to die on this hill.

Open ##5014420
@BioMan@awful.systems 2026-10-13 19:37

Self-replicating prompt text is a much more interesting vector for misbehaving AI systems

Open ##5018280
@BioMan@awful.systems 2026-10-13 19:38

Self-replicating prompt-injection text is a much more interesting and much more immanent threat to cause AI systems to misbehave in a propagating way

Open ##5018295
@istewart@awful.systems 2026-10-13 23:22

I wonder what answers you would get if you polled these Anthropic people about the Covid lab leak theory

Open ##5021069
@BurgersMcSlopshot@awful.systems 2026-10-14 12:04

I too have seen the documentary “Maximum Overdrive”

Open ##5029201