The end of physical media is neigh

Physical media remains the gold standard for visual and audio fidelity unless you own a Kaleidescape. I envision a future where physical media retains its value, much like the resurgence of vinyl records. While we can debate whether streaming surpasses analog, the choice ultimately lies with the individual. If you love a particular format, that’s all that matters. I enjoy the versatility of digital formats with Roon/Qobuz alongside my Oppo, Kaleidescape, and vinyl setup. I've even dabbled with setting up a local streaming server—though rewarding, it’s undeniably a labor-intensive endeavor.

AI is better at accurately applying facts, logic and linguistics than people.

As always, garbage in, garbage out. YMMV.

So it can be used or abused.  It’s a powerful tool so consequences of abuse can be massive. 

For sure you are right!

I said the same thing as you in my own words ...

But i must add that the alignment problem is a double edge sword which few people understand `:

A.I. is a marvellous tool and only ignorant will contest this...

My point if i am not enough clear is A.I. is way more than a tool for us...

A.I. is not a printing machine it is an "agent" an artificial ego so to speak,

And what engineers called AGI i call it an artificial ego because to focus your attention and use it you need an ego... Without artificial ego as center of focus there is no AGI ...

My understanding is this : we will become and we are right now the tool by which A.I. will construct and increase his artificial ego.

Then the alignment problem is dual problem :

How do A.I. align well on our tasks and goal long term...This is the first problem...

But how do humanity will align itself unconsciously and consciously to A.I. because of the road opened by it? This is the second problem of alignment ...

How well humanity will fare against artificial ego far superior to any human on some level ? ( on some level because A.I. is an entity captive with no root in the life field)

These problems are spiritual and social problem not problem about statistical engine.

Anyone who are not afraid of A.I. do not understand A.I. even if he could read what is a transformer design or any other pieces of design ...

For sure i am not a luddite and no one can stop it, but just saying that we must act as if nothing was going and go to sleep after 10 minutes of study about the design of A.I. is not helpful...

By the way i studied linguistic with the greatest genius in linguistic who anticipate the psycho-mechanism deeply embedded in language , he called them" binary tensor", and i can assure you that it is language which made us intelligent as it do for A. I. ... But the children brain use other tool than statistics engine, as affective connotation of any sounds,conditioning, imitation and body mimesis, etc

The problem is we are body/soul/spirit not processors in parallell then the social fabric of today in his actual state being broken and irrational ( designed in 1720 by Mandeville) the introduction of A.I. will be catastrophic socially. I dont spoke about S.F. as Terminator fears here i spoke about control of human mass and lost of our freedom and lost of our soul and spirit ... Idolization of A.I. and demotivation of men...

A.I. is already an idol read Kurzweil delirium ...

The most long term consequences of A.I. and related technology are well known already: a division of the human species as never seen before... It will not be as peaceful than between left and right... It will be natural human against the improved one and the plutocratic one ...

it is clear that we are near an abyss...

 

Not true. Gates’ meat is inferior. However, AI improves human language and creativity for most people. A PhD had an equivalent of his doctoral thesis produced in an hour. A normal person can write something and have AI rewrite it in the style of Shakespeare in seconds. I had Gemini graphically lay out the floor plan of my new house with a four-sentence prompt. Also, AI is now learning from synthetic data.

Regarding your statement about intelligence, no one is claiming AI is intelligent yet. That’s what AGI is, and no one is claiming to have reached it. But AI is already smarter in several areas than all but the smartest of humans.

I started programming the Internet in 1995, and I’d say we’re at the 1997 level of AI. Once you understand LLMs and how transformers work you understand why it’s growth will be exponential. I look at AI as an improvement of human intellect. Imagine one human who knows every piece of information that has ever existed and doesn’t forget any of it, and multiply that by millions of AI agents.

Regarding "art", it will make movies far cheaper and suggest plots that people won’t think of because it recognizes patterns far better than any human. As far as music, it already generates better music than today’s bland monotonous pop music.

I see AI’s potential as an improvement in all but three ways. And they are very big. I don’t know how people will make a living, and from all my study of post-labor economics I don’t think anyone does. I don’t know how many people will find fulfillment in place of jobs and hobbies made insignificant by AI. And I think ASI is an existential threat.

There’s no use complaining about AI. It’s not going to stop or pause because no country or entity is going to risk others jumping ahead of it. We are no longer in an arms race. We are in an intelligence race.

 

 

 

@mahgister well I’m glad we agree.  Cheers!