Artificial intelligence and fetish

AI, artificial intelligence, the mainstream has learned about neural networks and is keenly interested in matrix calculations. You can’t hear it anymore. Like any sow put through the ring, at least the world will end, if not the Nazis from the South Pole will have their hands in it. We need laws! We don’t know what AI is or how it works, but the digital overlord has said there is intelligence involved and since that’s not our area of expertise, we have to believe it.

There is talk of a turning point (Scholzen’s revanchism and militarism is only a turning point for the SPD, a return to the past), there is fear of machines taking over, they are everywhere! AI Nazis and man-eating Terminators everywhere. Only a strong nationalist state and the Umbrella Corp. and Skynet and Bill, the Bill will save us!

Since neural networks have become known to the general public through ChatGPT and image generation, it is not all that surprising that intelligence is easily believed to be artificial. After all, it certainly gives the impression of interacting with a human being, i.e. an intelligent being. However, these are only the spectacular use cases, the pop culture of self-referential systems, so to speak.

What we understand as AI today is anything but intelligent (not even ChatGPT). An AI is a statistical tool for processing Excel spreadsheets. Pure math, evaluating databases and calculating probabilities. Such an AI can never become intelligent, so it is completely wrong to call it that. When an AI writes a text message, it does not understand the content. It doesn’t understand anything at all, an arithmetic operation, a mathematical formula has no consciousness. If I calculate a triangle using the Pythagorean theorem, the formula knows nothing about triangles. Anyone who thinks that they know something about triangles and even have an opinion about triangles would rightly be dismissed as a crazy crank. It’s exactly the same with matrix arithmetic. It only “knows” the statistical probability with which a word will next be perceived as correct by the reader. And only from the tables provided by other people. An AI never reinvents anything! It can’t, if it’s only about probabilities from existing and processed data.

For example, an AI will only generate a correct image of a person by chance. It simply has no “imagination”, no concept or idea of a person. It can only calculate how likely the following pixel arrangement is to be perceived as a human (you, you learned this by clicking on the image you thought was correct in Google Image Search). Nothing about it is intelligent. Nothing.

An AI only becomes dangerous, like any tool, in the hands of a human. I can smash heads or build houses with a hammer. You know that. That’s why those who use this tool against you (let’s call them GAFAM) want you to be afraid of AI and think it’s intelligent (because of this misleading name). Not that you realize that the real danger, as always, comes from an imbalance in society. The fight for AI is also class warfare (yes, I know, nobody wants to hear it).

The people who control the tools have the power. The tools themselves have no power.