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.
A bit of clarification for people who are OotL (simplified from a few courses on ML, but maybe helpful):
The reason generative AI has exploded is because of the discovery of the transformer architecture in 2017 which itself was an innovation on deep learning. Transformers are special because they enable extreme parallelization of tasks that used to be heavily serialized (namely ones that needed recurrent neural networks (RNNs) or their more advanced variants like LSTMs).
We aren’t starting to mix this up with “classical machine learning”, because classical ML isn’t about the task; it’s about how you’re performing it. ICR, voice recognition, etc. in 2026 are largely done using transformer models. Even those ML tasks that aren’t using transformers (which do have their own downsides) are usually still using a deep learning model (for example, KDE’s Spectacle screenshot tool has OCR, and I’m pretty sure they went with an LSTM model). And it’s also why separating it into “GPU or no GPU” is a wild oversimplification (classical ML models, such as SVMs, can also take advantage of parallelization in training).
Classical ML (supervised and unsupervised) are techniques and architectures like SVM, k-means, decision trees, etc. Deep learning incorporates these classical ML concepts (for example, the most basic type of deep learning model, a deep neural net (DNN), is also a “multilayer perceptron” – where a single-layer perceptron is a bog-standard linear classifier and the granddaddy of ML models). But still, “classical ML” in the field is “machine learning that isn’t deep learning”.
If you want to learn how these models work and don’t want to be smacked in the face with textbook-style math, the YouTube channel StatQuest gets a lot of students through their ML courses. And if you want to try hands-on, there’s an extremely user-friendly Python library called
scikit-learnthat’ll be more beginner-friendly than e.g. PyTorch or TensorFlow.Thanks for the explainer.
Yes I oversimplified a lot, as I think the audience for this post are people that are not technically deep into the topic.
Maybe a post about something like from LeNet to LLM could be something in between deep knowledge and current hype. Just not sure where
A problem i had with your post (which this comment addresses) is that things like voice recognition and image recognition which didn’t used to use generative ai now often do use them on the back end.
So “AI” has become a synonym for the type of AI we don’t like, and machine learning “ML” for the limited tasks which we do like classical AI to do, like optical character recognition (OCR).