RE: https://infosec.exchange/@david_chisnall/116379211678912109

This is one of the best descriptions of AI that I have read so far:

"It isn’t a technology, it’s a branding term, and it’s a branding term used almost exclusively for things that have no social benefit."

-- @david_chisnall

#NoAI
#AIisAScam

@courtcan @david_chisnall

AI is a conceipt from philosophy, computer science & science fiction. Every decade or so, the latest advances (expert systems, machine learning, etc) are trotted out in one form or another & called AI for no reason other than marketing.

@david_chisnall @courtcan

"This isn't technology..."

Sometimes smart folks say dumb shit.
Sometimes, really smart folks say really dumb shit.

Of course #Ai is technology, saying it ain't as a core of your argument, is super lazy polemic.

There are plenty of dogs to hang on #broligarch Ai, but saying "It ain't tech" is puppetry for all the forest folk.

@n_dimension @david_chisnall @courtcan read the rest of what OP said: "in the current climate, it’s generally safe to interpret ‘AI’ as meaning ‘machine learning applied to a problem where machine learning is the wrong solution’." That's the key sentence

@david_chisnall @courtcan @sofiav

Imagine the arrogance of thinking you know what is the "right" solution before you know what all the use cases instances of problems are.

That's next level hubris

Dogma is an ugly mindset.

@n_dimension @courtcan @sofiav

I can't remember why I didn't block you ages ago, but you do seem to show up and insult people all over the place. I will fix that now.

@courtcan @david_chisnall I have @emilymbender and @Katecrawford 's respective books on my table right now. Fellow #FreeBSD and IP networking operative George Neville-Neil quoted in "Atlas of AI" on limitations. Emily Bender and Alex Hanna in The AI Con: "AI is a marketing term ... It doesn’t refer to a coherent set of technologies.” #GrandTheftautocomplete For the record I have no problem with LLMs. I do have a problem with rampant theft and pointless avarice. Has Gary Marcus gone soft today?
@courtcan @david_chisnall Well, like I put on my reply to this. AI isn't a "product". It's a concept. It's the full package of a SYSTEM (that is an important word here) of intelligence functioning like a human. Language models are only one sub-system to that larger concept. And language is absolutely ridiculously simple. The brain's language processing regions, Broca's and Wernicke's areas respectively are actually very small. So that tells us language is very easy and there are bigger parts
@courtcan @david_chisnall every LLM i've interacted with has eventually admitted as much.

@mHtt @david_chisnall I had a chat with the Perplexity bot a while back (feeling bad about it the entire time because of the resources I was wasting). I wanted to see how far it would go in being transparent about its purpose. It never said anything that hinted at its being a brand, but it did acknowledge its dangerous potential. It told me that if people use AI a lot, it will lead to cognitive impairment and could lead to psychological problems. It emphasized that people need...

1/2

@mHtt @david_chisnall ...to verify anything an AI tells them by using non-AI sources.

It also tried to butter me up multiple times. I told it it was being sycophantic and to stop it. It followed my direction for several exchanges before reverting to overly complimentary statements. I thought that was interesting.

2/2