Wait how are we defining most? I would imagine most models out there aren’t flagship models with billions backing them or with data centers. Most models seem to be specific domain small training models made with somewhat insignificant resources in comparison to flagship models.

Some AI, not all. I think it’s fair to make a distinction between the two. There’s very useful models with a smaller scope that aren’t part of these issues.

In 2000 the whole data hording apparatus hadn’t been fully implemented yet, and AI surveillance was something more so seen as science fiction, nothing to worry about until possibly 2100. Sure there was some data collection but it wasn’t like billions of people were walking around with surveillance tech in their pocket 24/7. Things have changed and drastically. Methods to monitor, model, predict, and control human behavior are now a thing and quite robust. If only people were aware of what’s happening there would be a mass migration to secure hardware, OS’s, comms, and of course not a single cloud based always on assistant in sight. IOT devices are cool when they are secure but they often get the least attention, few if any security updates. If you can get rid of robot vacuumes, smart home assistants, smart TVs, smart thermostats, switch to secure FOSS OS’s on your computer and phone.

You can point a laser at someone’s window and accurately record their keystrokes from the sound vibrations. Things go deep in the surveillance and intelligence world. Nothing is truly and absolutely secure, just more difficult to crack.

Sometimes there’s a final spice level called “Indian hot”

So chrome OS but Microsoft version? As a Linux only user I feel for you guys. Windows 7 and XP seem kinda revolutionary now in comparison.

Honestly we just need to divide the internet into three sections. The human only content and posts section, the human and AI section, and the AI mainly section.

Alopecia of the armpits?

There’s some loopholes to jump through depending on hardware but that’s also because of windows.

Kinda reminds me of WWII when we had Japanese American soldiers and likely their families in camps with all of their assets seized.

How about we don’t duplicate George Orwells 1984 irl?

How would you prefer people talk technology that allows machines to think, learn, and perform tasks that usually require human intelligence?

We could encode some behavior with mechanical automaton but we can encode far more behavior and more complex behavior with electronic computers. There’s no known strict limit beyond computation to the complexity of behavior we can encode. Now with a fully digitized fruit fly brain one can theorize about a fully digitized human brain. What would we call this if not artificial intelligence? How about other models that encode approximate behavior to human level intelligence?

The term has linguistic meaning and isn’t just marketing nonsense like calling food products all natural. That said there’s definitely been a ramp up of people calling essentially if-else chains artificial intelligence, lots of abuse of the term.

Sure, you know the term existed long before it became a financial buzz word right? Anyways in which sense are LLMs or NNs fundamentally flawed? I still believe your chief complaint is about the actions of people and not some mathematical structure much in the same way arguments about engagement algorithms on social media platforms is more about people and their choices than C++ or a binary tree.

I’m not saying the use cases that some have chosen for certain tools is neutral. Their behavior is indeed bad and should be called out. Ai for instance is at it’s core just a mathematical structure that has no intrinsic behavior, same for the concept of algorithms for social media or algorithms for drones or algorithms for the internet. Yeah some people use the internet to cause a lot of harm or even guide the outcomes of nations but is it the fault of html, CSS, and JavaScript? Not really. The problem is some companies, some people, some government officials and their choices.

You give them to someone who can or like in my situation where I’m aware eventually I’ll be traveling constantly I just don’t keep pets.

Or if you can’t take care of pets properly just don’t have them?

So which algorithm is that? If you’re hinting at manipulating the populace because in fact they are easily programmed via social network manipulation. Yeah anything done maliciously is of course bad but I think that comes down to the use of the tools and not the intrinsic nature of the tools themselves. Algorithms are just math, it’s the will of the individuals wielding them that is corrupt and that seems to be your actual chief complaint.

That all depends on how it was made. Not all models cost that much. Chinese models were mostly produced for far less mostly.

It’s called marketing. Somehow people are aware but still fall victim to it

Like which ones and why?

Why? Do you have that feeling towards other algorithms?