In the last weeks, I have seen some posts against kinds of AI, we are using as a society since decades (yes, decades) just because the Term AI is used.
When we are talking about society and planet destroying AI, we are talking generative AI. That is that shit we know als Claude and ChatGPT.
But media (and we) are starting to mix this up with “classical Machine Learning”
Some stuff that is AI, and I am pretty sure we are not fighting against:
- hand written numbers recognition (developed around 30 years ago)
- lane hold assist systems
- voice recognition
- image classification (computer aided diagnostics in radiology is around since 2010).
Those tasks do not need GPUs, huge amount of RAM or electricity.
We should differentiate between Generative AI (LLM, Video and image Generation) and those old school, valid and useful “AI” systems.
Not seeing a difference between those things is a AI Bro behavior, and should be beneath us.
Just being against everything because somebody wrote “AI” besides it only offers AI Bros an attack surface on every other argument we bring forward.
Everything that makes GenAI bad (hallucinations, AI psychosis, energy consumption, RAM usage,…) is a result of the generative part.
An AI model that recognizes YOUR (yes, your!) handwriting on a envelope and routes your letter correctly, and does that since decades, is also AI. I hope nobody here expects us to go back to sorting boxes and hand sorting of letters and packages.
EDIT: for more detail on AGI vs. GenAI look at this great comment.
Umm, no, AGI, or Artificial General Intellegence, the new term they invented for actual AI, doesn’t exist. And it’s very likely a pipe dream for all we know.
No, what is currently marketed as AI, including Claude and ChatGPT and all the others, is just an LLM. On steroids, sure - but just a generative machine learning algorithm. There has been no breakthrough; that’s why they’re desperately making the LLMs just bigger and bigger, cause they got nothing else.
Yeah, we need to be clear on taxonomy. Just about everything we hate about “AI” is LLM shit, which gives the illusion of intelligence but has absolutely NONE. It’s a parlor trick.
Generative =/= General
I did not use the term AGI? Generative is not a typo and the term I know for AI the generates content instead of just acting on it.
And to create something like AGI, we would need to adopt different ways to train a Model, like World Models. But that would get into deep AI theroy fast. And I would think that would be out of place here.
Ah, I realize where I misunderstood. My bad.
Still, though, isn’t every “AI” algorithm they’ve come up with a generative machine learning or what have you? Cause now I’m confused 😝
Yes, that is confusing.
LeNet, the oldest neural network I can name out of my head, was designed to recognize ZIP codes. 1990ish. It is the grandmother of ChatGPT. It does not generate new content. It uses an input set and translates it. Same for image classification in general. We take an input and translate it into a bunch of classes.
An LLM expands it’s input. So it generates NEW content.
And LLM is large language model? As opposed to a “language model”, which I figure is the same mathematical principles but smaller scale?
A language model is any model that operates on language in general. They vary in capabilities. Language recognition is one example of a language usecase.
Large language models is the class of models that are used for ChatGPT and other modern chatbots.
OK, that part was simple enough. And yeah, language involves some fascinating math, actually, so it kinda makes sense as a category for generative or predictive algorithms or what have you.