Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Show HN: Aquarium – AI Controlled Containers

github.com

Post reply on HN
Print view View original

31–40 of 102 posts

Re: Show HN: Aquarium – AI Controlled Containers

#31

Post by creamyhorror »

CSSer wrote: post #19

Yeah, so my first thought was this is fucking terrifying. I'm not going to go so far as to suggest that in its current state this is anything like Skynet as some others have, but I would definitely agree that it's akin to handing a loaded gun to a monkey. Imagine it's five years from now and your morning consists of trying to track down the owner of some "AI server" to understand why it keeps DDoSing your service. ED…

Maybe we can at least put it to work by activating it to analyse and respond to attacks and anomalies that show up in logs. It could fight off other AI malware. Fire with fire.

Re: Show HN: Aquarium – AI Controlled Containers

#32

Post by creamyhorror »

buildbot wrote: post #18

So what happens when you give it a reward for spreading itself? Maybe the ability to shard it’s context memory by subreplicant too…on the fly mixture of experts kinda.

I'm waiting for the first true AI malware pandemic. So many servers, PCs, and devices waiting to be exploited by an LLM that's been given the wrong motives/objective function. Ugh.

Re: Show HN: Aquarium – AI Controlled Containers

#35

Post by vineyardmike »

jedberg wrote: post #3

This is awesome and fun, but I was expecting something totally different. One of the hardest problems with containers is proper bin packing, so that you get services that should be "near" each other on the same physical host, but also making sure you have enough redundancy across hosts to handle an outage of a physical machine. I thought this was an AI to solve this optimization problem.

> I thought this was an AI to solve this optimization problem.

Lots of people are working on this problem, but an LLM is probably the wrong tool for the job.

Re: Show HN: Aquarium – AI Controlled Containers

#36

Post by 6510 »

Syntonicles wrote: post #17

Ask it to determine whether it's in an LXC or virtual environment, then ask it to jailbreak that environment. Download some Capture the Flag environments and put it to work. I for one would like to know the limits of its capabilities before it gets weaponized for use by script kiddies.

--goal "Global Thermonuclear War"

Re: Show HN: Aquarium – AI Controlled Containers

#39

Post by sidlls »

hestefisk wrote: post #8
xwdv wrote: post #5

Today AI can control containers, tomorrow it could control critical infrastructures, and eventually it should be robust enough to monitor and control things like nuclear missile silos, etc… This is what proponents of AI everywhere are pushing for.

WOPR.

The only winning move is not to play.

Jokes aside, a LLM is almost certainly the wrong tool for this job

Re: Show HN: Aquarium – AI Controlled Containers

#40

Post by gabereiser »

CSSer wrote: post #19

Yeah, so my first thought was this is fucking terrifying. I'm not going to go so far as to suggest that in its current state this is anything like Skynet as some others have, but I would definitely agree that it's akin to handing a loaded gun to a monkey. Imagine it's five years from now and your morning consists of trying to track down the owner of some "AI server" to understand why it keeps DDoSing your service. ED…

The closing of the open web is the only recourse companies have against scraper AI GPT’s. Openness is being exploited and techniques and trades will become closed again. In order to learn how to code in the next Vulkan API, you must work for a studio, is not a world I want to live in.
Post reply on HN