So my 5 cents - I work in IT as a programmer and have for the last 10 years or so. I'm no genius, most likely a moron like most people but I have a basic understanding of how this technology works and tried if a few times.
Its mostly just hype. The tech itself is just very good autocomplete and the name "AI" iself is an oxymoron. There is no intelligence.
Dont get me wrong, its cool and even useful in some cases, but the current approach of this tech will never be able to solve some problems we find easy because its just autocomplete.
There are a few simple examples like ask it to draw clock with certain time - it often fails and draws 10:10 instead of the time you asked because thats what most sample data had. Ask it to solve math problems or count letters in a word and it fails too. Thats because as I said there is no intelligence, it doesnt understand time, clocks, math or numbers. It just trying to predict which autocomplete option will statistically produce the best results.
I'm not even remotely worried about my job because even with all the type information and context from syntax programming has it still fails to provide functioning code in like half the cases even it its just a single line of code. This eventually made me stop using it because it looked ok at a glance but it used the wrong variable name or so which is super easy to miss. Any social media post you see about how its awesome and did X or Y project with 0 coding knowledge is bullshiting you. If you try it yourself you'll quickly see that anything beyond a single web page with text will be too complex for the "AI" and it will fail to produce anything useable. Or just take a look at how "AI" is playing pokemon, a game made for children that are starting to learn to read and is still failing to progress.
The few use cases I see is partial replacement for search engines, especially cases where you dont know the exact search terms, or in disposable arts. Stuff like generating backgrounds for ads or generating the ad text are perfect examples of tasks the currrent generative tech is good at. Art that has basically 0 artistic value and needs to be mass produced. So for example I ask it to write emails for me by just giving a short outline and the general tone I want to convey. Perhaps even some automated data processing where accuracy is not an issue. In general, anything where the "solution" is not 100% exact - asking a solution to math equation has one exact solution but if you ask it to draw a bluejay on a windowsill theres infinite solutions.
Also all the stories about it trying to break free are absolute nonsense. Even if it was somehow possible for the "AI" to come to this conclusion it takes massive computing power and storage space for it to even exist, like you most likely wouldnt be able to run it locally, theres nowhere for it to escape to.
The scary thing about this "AI" is that guillible and/or greedy people will try to push it into areas where it shouldnt be used, like some very important systems such as electric plants, police or courts and such. I saw some people already mention war and warfare - can you imagina an autonomous drone with a bomb that hallucinates and just turns around on its owners or something like that? Or an automated accounting system that simply doesnt understand numbers? Thats just disaster waiting to happen.
