Misconceptions about LLMs & the real AI revolution

09.11.2025      LinkedIN      Reddit      

Before everything has been AI, things were "smart" and before that "digital". With smart things like smart phones I never really felt
like they were smart. They often merely had a couple of algorithms to make things more accessible, often poorly executed to slap
the next buzzword on a product. Since then, it seems the tech industry is ahead of itself with this framing. The same goes for AI.
Now bear with me, it's going to get philosophical.

After ChatGPT-4o, I have to admit it caught me off guard for a moment thinking big changes are ahead. They very well are, just not
with the current approach. And this is the problem with the here and now. A lot of funding, private and tax payer money is impacting
our lives in many ways and lead into - what I believe - is a dead end. Although the current quote on quote "AI" is solving real problems
and it is nice to quickly generate an image for a blog article, it is not the AI revolution people expect. Here is why not.

Imagine a network of probabilities - an arbitrary system of causally connected nodes - is able to develop a consciousness. This would
in turn mean, that any system of causally connected nodes can be a conscious entity. That means, any superset of system of causally
connected nodes can be a conscious entity. And that means inside of you countless conscious entities exist at the same time, each
believing they are alone in there having original thoughts. The same would go for any material thing, really, because everything is full
of connected nodes in different scales. It can be molecules, atoms, quarks, but also star systems and ecosystem each being a
conscious entity. I do not know about you, but for me this is breaking reality. And just imagine what you are doing to your are doing to your
toilet brush everyday!

Let's take it further. If LLMs and other material things can not become conscious by being a complex enough system, that means our
consciousness is not material. Do not take it as god-proof, though (looking in your direction, religious fundamentalists).

What I am saying is, that the current state of the AI industry will change again and the software stacks as well as the hardware around it
will be in far less demand. The real AI revolution will not be consciousness, I think. My belief is, that the revolution lies ahead with
insanely efficient memristor chips so that everybody gets to have his own little assistant. I am not so sure about general purpose
robots. The complexity of the outside world has not really been managed to deal with without even a glimpse of light in there, which
even goes for plants, and ants.

I want to end this with some food for thought. If we some day can definitely confirm to have created a consciousness, we may suddenly have
cracked understanding of ourselves in such a profound way, that we turn away from hype, misery and infancy of our species. One more thing
though: upload you into a machine can never keep you alive. You would vanish as the wonderful conscious entity you are.

Stay optimistic and don't get caught in the noise of hype and echo chambers.
Cheers

Wir benötigen Ihre Zustimmung zum Laden der Übersetzungen

Wir nutzen einen Drittanbieter-Service, um den Inhalt der Website zu übersetzen, der möglicherweise Daten über Ihre Aktivitäten sammelt. Bitte überprüfen Sie die Details in der Datenschutzerklärung und akzeptieren Sie den Dienst, um die Übersetzungen zu sehen.