The LLM I use is fairly quick to retreat into extra disclaimers. "Not sure about this, this is from memory and you ought to check", etc. Don't they all do that? Is it just because of my prompt (which does say to push back if I appear to have bad ideas)?
It seems possible that people tune out disclaimers and caveats in the middle of the answer, too.
It formats mail flawlessly, signs as necessary by default, sets up every detail and checks that it's correct during installation, what more do you want?
You can heat with coal if you search long enough, too. There might not be a retailer around the corner from wherever you are, but the number of coal retailers isn't zero.
In this case GP probably knows that e.g. Italy has a life expectancy well above that of the US. Still, when he read that well-off people in the US have much better health coverage and it takes months to see a doctor in Italy, he believed it. How is it even possible to believe such a thing when you know the numbers?
You read that? I can get either in a day or two if I need it, they want much longer-term planning though.
For a vaccination I'll call several weeks in advance. If I want an appointment before/after work I'll call well in advance, if I want something tomorrow they may say "11:30, take it out leave it", which isn't great for my work.
Don't trust whoever wrote what you read there. FUD scaremongers.
I've done something like that too, but I find it restricting.
I want back-and-forth, very approximately like when I do pair programming. The dividing line between what I do and what the AI does varies according to task and sometimes during the task, and is seldom clear at the start.
Then there's the work that wants a human to click buttons and decide whether something is a good and correct user experience. The AI does not have access to my display if I can avoid it.
Overall, the model you describe is one that's worked very well for me, but for some problems. An unsatisfyingly small set.
"let's send this stuff from the warehouse to Ukraine" "let's say it's a value of $x even though it's close to the use-by date" and then come you and write that someone sent $x, and expect anyone to believe that.
And you even write that $x makes hard things easy. Hope much did the US pay to lose against Iran? Winning against anyone for the same sum isn't easy.
Sometime who worked for an advertisement provider told me it worked well as a signal, though. This was before it was obsolete. So few people used it that it helped identify its users.
(His employer explicitly ignored it. They used an ML algorithm to assess signals, but they overrode that particular assessment as soon as they discovered it.)
Why is it going to be huge? Who are the customers and what do they want?
I don't follow this closely, just look at the pretty pictures. If there's demand for lifting much bigger/heavier things to orbit than presently possible, I would probably not know, for lack of pretty pictures. So please tell.
A small town where I've lived was very much like its neighbours, but one particular neigbour was different in two clearly visible ways: ① there were (still are) more rich people in that neighbour and ② it was much easier to get financing for starting and growing companies in that neighbour.
I live in Munich. Couldn't find numbers for the metro area, but I found numbers for Bavaria. It seems that Google's electricity usage has grown to the point where it uses more than half as much energy as the Bavarian car-using commuters do. Still less, but more than half. (This counts only trips between home and work and only those by car, and the number is decidedly imprecise.)
So at this point, I assume that Google uses more power than the Munich metro-area single-car commuters, even for the widest definition of metro area (the widest I've heard spans 6m people, four times Munich's own population).
Is this reasonable? Is it reasonable that a collection of services with >1B users uses more energy than the single-car commuters in a metro area like Munich? I rather think so. (But then I don't have a high opinion of the value of driving to work.)
Enshittification is neither invariant nor universal — Rolls Royce has built cars for a hundred years and the new models still aren't Ladas or supermarket shopping carts.
Claude is a paid service, not one of those freebies where you are the product and receive as little consideration as practically possible, while advertisers are the paying customers for whom the service is optimised.
We're still arguing for several reasons, one of them is that people still confuse the user with the owner, as you do. "The user must be able to override" is implies that if you have physical access to someone's phone, you can install a keylogger before handing the phone back its owner. Nice for you but I imagine the owner might still quibble, even if you quote TRON.