Soon it won't even benefit the billionaire pedophiles who pushed for AI in the first place. Data centres cost billions and put already underpaid employees out of work, meaning more and more people won't be able to even buy from the corporations who shoved AI up our asses, causing a bigger drop in profits. AI has become a lose lose idea.

Disliking AI is completely acceptable and more than fair.  Using fictional characters as the basis of an argument is not.

I am happy to have an intellectual discussion, but it can't start with made up stories from Hollywood.  

Imagine a different situation where my proof that we don't need to worry about global warming was because the Na'vi 12 ft smurfs are going to save our planet.  Ridiculous, yeah?  

I wouldn't use Skynet as an analogy personally; as I understand it, AI in its current form is unlikely to become the sort of malevolent superintelligence depicted in films like The Terminator.

What worries me more is issues like the enshittification of the internet and economic fallout.

Frankly I think AI is a buzzword, it isnt AI.  It is a digital library that be searched at lighting speeds.  It is basically google chrome on steroids.  It isnt thinking.  

What AI does to jobs is flatly terrifying.  What used to take a week in analysis can be done in 30 minutes.  That is going to be a massive problem.  

Yeah the name is a marketing ploy; it doesn't actually "think" or have any real awareness or intelligence.

The problem with AI (one of them at least) is that it hallucinates and makes mistakes; I've read quite a few companies that were eager to replace all their workers have found this out the hard way and have ended up having to re-hire people to fix all the AI's mistakes.

It's ultimately a tool, but as a tool it can be misused by those who want to cut corners to save money.

Exactly, which is why the whole Skynet and sentient crap was stupid.  Basing AI is fine, but it should be grounded in reality with some level of actual knowledge. 

Copilot can do some extreme maths very quickly (and accurately).  That is my primary use.  

It just makes everything faster.  And the problem is, if China goes all in, the rest of the world doesn't have a choice.  That isn't how a global economy works.  

This is a common misconception, but there's no direct storage of text whatsoever in language models. The average parameter holds just a few bits of data compressed to hell and beyond, you can't build raw text chunks out of that. Language models can spit out verbatim quotes, but that's limited to heavily repeated (just like rote learning memorization works for humans) or very unique sequences in the training data.

As for whether that's AI or not, let's just remember that both the Eliza effect and the AI effect are real things. Depending on their bias and their priors, people are just as likely to anthropomorphize something that isn't really feeling anything as they are to discount something as "not real intelligence" once computers are able to do it, while moving to their next permanent bastion of human supremacy.

Rationality doesn't take the personal experience into account. It's very rational to say the world would function better with far less people, release Covid 2.0. Rationality also argues torture for information to prevent worse is acceptable, imagined or not.
Rationality is also excluding deviations from the group to optimize the functionality of the average.

I fear rationality! Rationality without humanism and empathy.

Current AI is also a reflection of us. It's an accelerationist factor on the terrible path humanity is on. AI will only increase inequality, not solve it. AI will also make things less safe when more people rely on it, use it as a 'shortcut' and forego understanding and deep knowledge of complex system for higher productivity. As well as deflecting accountability, the machine did it.

This is unrelated to Copilot but I wanna share this AI news here anyway since it's a huge deal. It'll depend on how much things continue to progress in the coming years but it's looking very possible to me that AI revolutionizing the field of mathematics is right around the corner now which has massive implications. 

Still more realistic and with a better outcome than what the SpaceX IPO will become.

I mean, over 28 Trillions (Yes, with a T) in TAM (Total Addressable Market), most of it for... AI:

Yep, Elmo unironically puts Grok's addressable market as 26.5 Trillions, while Space-related stuff could be happy to reach 2 Trillion added together.

Of course, the ridicule on Ars technica for those numbers was almost as large as that incredulous number: https://arstechnica.com/space/2026/05/spacex-submits-detailed-financial-filing-ahead-of-going-public-in-june/#comments