I was born in the second half of the 20th century, I read Pinocchio as a child and the Grimm brothers and more. Those were the books for children. Did they damage us compared to kids of 50 years later? I can't tell. Probably nobody can tell until at least next century.
I don't know which version you read of course, but worth noting they've been increasingly diluted over the years, in much the same way as OP says Disney did to Pinocchio.
I would naively suppose that the agent is able to read the man page or run the help command of the tool. They usually contain plenty of information. But bending the tool to suit the agent has some value. The GNU-AI suite of userland tools? Unfortunately it's possible that every model will settle on a different average. If that's the case we can't bend to every model. Models will have to bend to whatever we want to use.
Of course it can read the man page and run cmd --help.
Now you've wasted context on, what? Learning how to use the tool. And it will waste context on it every single time. (You can write skills to mitigate this a bit, but still).
The alternative is to make the tool work as the user (an LLM in this case) expects it to work, without having to resort to the manual.
If the parameter names mostly standardize across tools because the models learn to predict those names, then humans will also learn to predict those flag names so this actually has the potential to make tools more human friendly and easier to learn.
There are many theories of consciousness but nobody knows if one of them is correct and nobody can use one of them to build a conscious machine. Compare that to theories of physics. None of them is 100% correct but they give us the tools we are using to write these messages.
I've got a theory of consciousness, not a very complicated one, that could be used in a machine. Basically that it evolved as a practical way for animals to make decisions like whether to run from a predator. To do that info from the billions of neurons handling senses memories and the like filter down to something like a situation summary, which is basically what the animal is conscious of which then feeds to the decision making, thinking and remembering and neurons.
It would be quite interesting if/when someone tries that to see how close it is or isn't to nature.
So, your theory says it's about prediction, estimation, extrapolation? Like a Kalman filter?
I'm thinking of the first episode of Alex Garland minisery Devs, where a somewhat brilliant Russian guy demos predicting the exact movements of a bacteria. That would be proof that the model works. And even it it were to mispredict, most of the time, it sounds plausible.
To be honest I'm not that up on those terms but I figure given that life evolved the functionality must be something like what I mentioned. I'll have a read up.
I don't know if it's an Android thing or only a Samsung thing, but my phone displays colored dots over the buttons that led to some new functionality. The dots go away when I eventually push those buttons. If I follow the trail of dots I can discover what's new. It's quite unobtrusive but pushes me to look at those functionality and get rid of the dots. Sometimes I do it in 5 seconds, sometimes in 5 months, but it happens.
I never use my numpad. I use the numbers in the top row of the keyboard.
I'd be super happy to yank my numpad out of my laptop, move the keyboard a little bit to the right and center align it with the center of the screen. My head would be centered with the middle of the screen too.
Unfortunately I had to settle with that keyboard because every other laptop was a worse tradeoff.
I still have to understand what my AI agents could do that I don't want to do myself. Buy stuff? No thanks, I want to see what I buy. I think that they are 99% a solution in search of a problem.
Same. Well the biggest thing I don't want to do that they could help with is work. But in the cases where it can do that for me, there's no world where that benefit goes to me rather than my employer.
Well, that's the very nature of the employer / employee relationship. In my case I write software for my customers and I trade time for money. If I use an AI to write code two times faster my daily rate doesn't double. However I can keep my costumers.
That's only another step in the path I experienced since the 80s, when I had to type every single character because there was no auto complete, no command line history, very few libraries. I was very good at writing trees, hash tables, linked lists and so was everybody else. Nobody would hire me if I were that slow at writing code today.
My family (unfortunately) uses InstaCart and probably 15% of items are a shitty "replacement" "not what I wanted". For time sensitive items, having the shitty replacement item NOW is better than having to wait for the "item I actually wanted", so we often just accept the inferior product. This is a dark pattern that I could see AI adopting -- it buys tons of cheap crap you didn't want, some of it was right, and you're left with a mess of returns to sort out, esp. if those returns require you to physically take some sort of action like physically returning the item to the store
I'm on my phone now so I can't check if something has changed, but what you want to protect from change is the directory, not the files. A file can be deleted and created again if the process can write the directory.
reply