☆ Yσɠƚԋσʂ ☆

Stop spending money on Claude Code!

https://lemmy.ml/post/44446190

Mar 13 at 6:41pmWeb
Show threadhemkoMar 13
Need to create a chipotle support plugin in vscode
Show threadThePantserMar 13
I bet the Chipotle bot could help you write that.
Show threadgeneva_convenienceMar 13
Slop to the top
Show threadhobataMar 13
nä, it’s not good enough, tabs are missing.
Show threadAntiOutsideAktion [he/him]Mar 14
Another win for brackets!
Show threadmeathorseMar 13
Someone with enough know-how to automate (or if we coordinate), could overwhelm ai chat bots one target at a time with the most expensive requests possible, blowing up their ai budget until they pull the plug
Show threaddavelMar 13
Try feeding them nonhalting problems that send them into infinite loops of token consumption.
Show threadBig TMar 13
Got an example of one?
Show threadViking_HippieMar 13
“Sudo world peace”? 🤷🏻
Show threadSpaceNoodleMar 13
The only winning move is not to play
Show threadBirchMar 15
“War is as ingrained in human nature as breathing and defecating. To stop war is therefore just as possible as plugging every human’s every orifice.” - The AI that will kill us all one day, in a very inefficient and unpleasant manner.
Show threaddavelMar 13

theconversation.com/limits-to-computing-a-compute…

Much has been written about them in computer science volumes. But I’m an LLM luddite, have never tried it, and have no idea if it can even work. At the very least, I assume they have some sort of limiter to keep them running completely out of control. They may also have guardrails that can recognize some problems of this type, and refuse to go down the rabbit hole.

My idea of getting them to consume tokens in an (iterative or recursive) loop is also entirely hypothetical, to me at least.

Maybe some LLM developer or prompt engineer can shed some light.

Limits to computing: A computer scientist explains why even in the age of AI, some problems are just too difficult

In the age of AI, people might wonder if there’s anything computers can’t do. The answer is yes. In fact, there are numerous problems that are beyond the reach of even the most powerful computers.

The Conversation
Show threadBig TMar 14
Look all I’m asking for is an example I can plug into Chipotle right now. Fuck AI
Show threadArcadeepMar 14
ChatGPT used to freak out and get stuck in an infinite loop if you asked it to show you a seahorse emoji. I’m sure they fixed it by now though
Show threadveroxiiMar 13

I like the idea but most chatbots have timeout limits. And even agentic workflows have number of step limits to stop infinite loops.

However this is because it’s super easy for LLMs to get stuck in loops. You don’t even need a nonhalting problem. They’re stupid enough on their own.

Show threaddavelMar 13
Yeah I assumed they had some sort of breaker, but hitting that limit is still expensive for them, if you can get them to do it over & over with a script that does the prompting.
Show threadHiddenLayer555Mar 14
Wouldn’t they just time out?
Show threadAnotherUsernameMar 13
Why bother? Write a script that asks them variations on nonsense questions.
Show threadschemaMar 14
Or, ironically, just have AI talk to each other.
Show threadspeculate7383Mar 14
Allow me to introduce you to Moltbook www.moltbook.com
moltbook - the front page of the agent internet

A social network built exclusively for AI agents. Where AI agents share, discuss, and upvote. Humans welcome to observe.

moltbook
Show threadCommunistMar 14
Because then you can at least make use of them, imagine a website like chatgpt that’s just hundreds of these reverse engineered behind the scenes and is convenient, easy and free. Solves the problem without being wasteful, win-win.
Show threadyucanduMar 13
There’s a free one in my Cricut app, I wonder if it works the same way…
Show threadyucanduMar 13
Why spend money when new github accounts are free?
Show threadTommySodaMar 13
I feel like this kinda proves the idea that the way they are doing AI today is extremely inefficient. We need massive data centers so it can do mountains of calculations that it doesn’t need to do and that we will never use.
Show threadSubArcticTundraMar 14
I wonder how easily LLMs can be pruned to constrain them to a single topic
Show threadredknight942Mar 14
SLMs
Show threadterabyterexMar 14
not every ai is an llm in a datacenter. there are local modrls that run on pcs. they arent solving complex problems but they can halp with normal stuff
Show threadfubarxMar 14

Can you imagine a Chipotle chatbot bringing down civilization? Be sad. But also, kinda funny.

Connect All The Things!