American cars have sucked since about the 1970s. They cannot compete, especially not the factories in the Rust Belt. So what do you do, if you can't compete? Easy: try to have your competition banned.

Some level of tariffs might make sense. An outright ban, no.

You're just lying. I ran some tests on the scenarios you described and the results were an absolute mess. They only worked in certain cases and would be practically impossible to maintain if necessary. What I found useful about LLMs was summarizing information and finding documentation (which is their strong suit), but writing code for you? And complex code? No way.

If your country lasts that long. And with the risk that he might decide "If I'm going down, I'm taking everyone with me" and end the entire world along with it.

As long as companies obtain legal access to material, I don't see the problem. Scraping? If a company puts material on the Internet, it can be seen. Scraping is not - and should never be - illegal. Torrenting is much the same: If the material is legally online, it doesn't matter how they accessed it. That assumes, of course, that they didn't seed any torrents.

What we have here, are lawyers looking for a payout.

Seriously? So which is faster: pressing a physical switch, or navigating through X menus on a touchscreen?

It's not just touchscreens, though. Our EV has the climate control functions on touch-sensitive "switches". Not part of the display, but nearly as bad, because there is zero tactile feedback. They are just part of a smooth panel, so you still have to look down to find the one you want.

It’s about control too. If all eletronic transactions have to go through VISA to be processed, then VISA can control who can buy what and who can sell what. A public, trustworthy eletronic payment system managed by other governments takes that power away from VISA, so naturally VISA will sabotage any attempt by foreign governments to establish something like UPI.

If I am presenting material I know well, I can "split" my attention, to do some quality control. One half of me goes on lecturing, while the other half is evaluating the presentation. Speaking clearly and loud enough? Too many ums and ahs? Looking at the students enough? Is the board work neat enough? Etc.

I don't do this often, because it is a really weird feeling and shifting back into just one "me" is a strange lurch.

I don't get it either. I use AI a lot, for all sorts of different things. In none of them do I want to talk to it. The only exception I could imagine would be if I wanted to brainstorm about something while driving.

Also, all of the voices have (to my ear) an annoying, fake-bubbly-happy California accent.

Every company ought to always be required to make technical documentation and replacement components available for their products.

I just had this on a much smaller scale: a cordless Bosch vacuum cleaner. It almost certainly needed a battery replacement. Nowhere on the internet is there a diagram of how to disassemble it. Remove all the visible screws and...nothing. Finally get it apart, and: the seven batteries (in series) are installed in such a way as to be very difficult to remove. Of course, there is no information on what their specs are, and as far as I can tell, Bosch won't sell you replacements. Ultimately, we threw it away and bought a new one (from a brand that does sell replacement parts).

It's absolutely irritating to have to trash an appliance that just needs a simple repair. This is where government regulation would actually be important.

Cool idea for satellites near the earth. The thrust will be pretty minimal, but since it's basically "free" that doesn't matter.

Moon and Mars...hmmm...gonna be a slight lack of magnetic fields to push against, as soon as you are away from the Earth. Neither the Moon nor Mars has a significant magnetic field.

ChatGPT points out that the magnets are likely more useful to maintain orientation. Any actual thrust will be on the order of nano-newtons. Let's be generous and assume an entire micronewton of thrust, and that this can be maintained throughout the entire orbit. For a 100kg satellite (that's quite small), that could give you a velocity change of 0.3 m/s over the course of a year. That could be useful for satellites in mid- to high-orbit. It is far too little to be useful in LEO. Of course, the magnetic field is weaker at higher altitudes, so...

Anyone with the info off the bottom of a check can withdraw money from your account. There is basically no security.

Join the modern world, folks.

You're just fooling yourself now... I've conducted experiments despite what you might think, and the results have always been so disastrous that I gave up trying. Same goes for my colleagues. And my interns may make mistakes just like AI, but they have one advantage: they learn from their mistakes.

Well, time will tell who is right... But the way things are going, my way of doing things may will end up being actually safer than this AI-generated slop that's popping up everywhere and being marketed as "the best thing in the world". And I don't have to worry about how many "tokens" I have to spend on what I'm doing, because, after all, I'm not using any.

You’re getting somewhat usable results from AI solely because someone else in the world has already done something exactly the same or very similar to what you wanted to do. This isn’t all that unlikely to happen because there aren’t that many different ways to make a form receive and display data, for example.

Now, if you take someone like me, who is constantly creating new things or new ways of doing things (and no, I won’t describe them because of trade secrets), AI just gets in the way and can even be a serious security risk. The only thing AI has been somewhat helpful with here is acting as a “Google summary,” and since I’m the one who decides whether the AI’s suggestion has merit or not, it never touches the code itself.

Cars are actually useful. AI, on the other hand...